Next Article in Journal
Light Stress Detection in Ficus elastica with Hyperspectral Indices
Next Article in Special Issue
Automated Windrow Profiling System in Mechanized Peanut Harvesting
Previous Article in Journal
Use of Phenomics in the Selection of UAV-Based Vegetation Indices and Prediction of Agronomic Traits in Soybean Subjected to Flooding
Previous Article in Special Issue
Optimizing Orchard Planting Efficiency with a GIS-Integrated Autonomous Soil-Drilling Robot
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Recent Advances in Agricultural Robots for Automated Weeding

HUMAIN-Lab, Democritus University of Thrace (DUTH), 65404 Kavala, Greece
*
Author to whom correspondence should be addressed.
AgriEngineering 2024, 6(3), 3279-3296; https://doi.org/10.3390/agriengineering6030187
Submission received: 27 August 2024 / Revised: 9 September 2024 / Accepted: 10 September 2024 / Published: 11 September 2024

Abstract

:
Weeds are one of the primary concerns in agriculture since they compete with crops for nutrients and water, and they also attract insects and pests and are, therefore, hindering crop yield. Moreover, seasonal labour shortages necessitate the automation of such agricultural tasks using machines. For this reason, advances in agricultural robotics have led to many attempts to produce autonomous machines that aim to address the task of weeding both effectively and efficiently. Some of these machines are implementing chemical-based weeding methods using herbicides. The challenge for these machines is the targeted delivery of the herbicide so that the environmental impact of the chemical is minimised. However, environmental concerns drive weeding robots away from herbicide use and increasingly utilise mechanical weeding tools or even laser-based devices. In this case, the challenge is the development and application of effective tools. This paper reviews the progress made in the field of weeding robots during the last decade. Trends during this period are identified, and the current state-of-the-art works are highlighted. Finally, the paper examines the areas where the current technological solutions are still lacking, and recommendations on future directions are made.

1. Introduction

Increasing food demand is a consequence of an increasing world population [1]. Some studies project that food demand can increase by up to 56% between 2010 and 2050 [2]. To compensate, a commensurate increase in food supply is required, and the agriculture industry is exploring new ways to realise such an increase. Labour shortages observed in many countries are hindering such efforts, especially regarding seasonal agricultural work. At the same time, the environmental impact of farming and other related agricultural operations also needs to be taken into account, and novel practices need to be adopted in order to reduce the carbon footprint of equipment and materials used in agricultural processes.
Agricultural robots have been proposed as, at least, part of the solution to the aforementioned problems. Automating agricultural tasks requires a level of autonomy so that human intervention is minimised. The implementation of such autonomous robotic solutions requires the integration of various fields such as machine vision, localisation and navigation, sensing and actuation, to name a few. Recent systematic studies have highlighted the progress made in developing robots and their supporting technologies that can automate agricultural tasks for the purposes of mitigating labour shortages, improving productivity and reducing the environmental impact of farming [3,4,5]. A number of solutions have been proposed for diverse tasks and diverse crops, such as monitoring vineyards [6], maise phenotyping [7], and strawberry harvesting [8]. Also, the proposed robots vary in physical characteristics depending on the target crop; they can be ground [9] or aerial vehicles [10], operating as a single unit or in a cooperative fashion [11,12].
One of the areas in which robots can play a role in automating is weeding. Weeds present a considerable challenge to farmers because as they grow, they compete with the actual crops for space, sunlight, water, etc. This results in reduced crop yields and reduced crop quality. As a consequence, weeds cause significant production and economic losses. Weed control methods are mainly chemical (herbicides) and mechanical (tools). Chemical weed control can result in herbicide-resistant weeds, which adversely affect crops and the environment. Herbicides are also subject to increasingly stringent regulatory restrictions. On the other hand, mechanical weed control tools are very crop-specific with variable effectiveness. Robotic weeders are a promising alternative to herbicides since they are less affected by environmental conditions, they are more cost-effective compared to the development of new specialised herbicides and pose less of a risk to the environment [13]. The challenge would, therefore, be developing effective weeding tools and supporting technologies [14] and focusing on improving weed detection algorithms [15,16,17] while at the same time mitigating the problem of labour shortages.
The present paper is a review of the research carried out in the area of robotic weeding during the last ten years. The focus of the paper is to examine the robotic platforms that have been proposed in the last decade and specifically to survey the various methodologies and supporting technologies they employ to achieve weeding, namely their physical characteristics, weed detection, weeding tools and navigation. The paper also presents some examples of current commercially available weeding robots. In addition, the reviewed works have been studied cumulatively in order to identify trends in the weeding robot research, that is, to identify where most of the research takes place geographically (in terms of geographical region as well as in terms of country), and to determine whether this area of research is becoming more popular over time.
The paper is laid out as follows: Initially, in Section 2. the methodology followed for compiling the complete literature on weeding robots is described. In Section 3, the research on weeding robots is analysed, including current commercially available systems. The observed trends from the literature review are also presented. In Section 4, an overview of the examined literature is given, and the various technological aspects are discussed. Finally, in Conclusions, the areas where further research is needed are identified, and proposals for further research are made.

2. Methodology

Several sources were used to gather the available literature regarding weeding robots, namely the databases of ScienceDirect, Google Scholar, Scopus, IEEE Xplorer, and Wiley. To filter the papers of interest, the criteria were that the publication date is from the year 2014 to today and that the paper involved a robotic weeding mobile platform. This means that works only presenting a supporting technology (for example, a machine vision algorithm for identifying weeds or an automated weeding tool) were not included. In addition, only the latest publication was used for research works presenting incremental improvements to the same robotic platform.
A total number of 48 research papers have been reviewed. Additionally, there were also 7 websites for commercial weeding robots. The pie chart of Figure 1 shows the number of robots reviewed per type of publication (journal article, conference proceedings, website, etc.).

3. Weeding Robots

In this section, the various approaches in the development of weeding robots are analysed. The reviewed research has been divided into two broad categories, i.e., robots which use chemical and non-chemical weed control methods. Chemical weed control methods involve the application of herbicides. Non-chemical weed control, on the other hand, involves mainly the use of mechanical weeding tools and laser-based tools. In each of the two categories, the papers have been examined according to their main focus, namely (a) locomotion and navigation, (b) sensing, and (c) weeding tool.

3.1. Weeding Robots in Research

3.1.1. Spraying Robots

The locomotion and navigation methods proposed in the literature depend on the mission (environment and target crops). Mainly, the robots follow a straight path within a crop row. For example, a low-cost spraying robot for crops with narrow rows is proposed in [18]. The robot is able to use vision to detect and apply herbicide to weeds, as well as to detect the crop row in order to perform navigation in-row. The proposed robot is able to operate continuously in flaxseed fields with an autonomous recharging system. The low-cost robot described in [19] is designed to move on a rail in pre-defined paths in sugar beet fields. In this study, the camera detects weeds and applies an amount of herbicide, which depends on the size of the detected weeds. The authors observed the decreasing amount of herbicide required while adjusting the speed of the robot and the height of the spraying nozzle.
In terms of sensing, robots predominantly use vision for weed and crop recognition and localisation. The purpose is to recognise the location of the weeds in order to apply the herbicide in a targeted manner, thus limiting the amount dispersed in the environment. An early work regarding the vision for precision spraying is presented in [20]. In this work, the cameras mounted on a modular robot are used to control the eight spraying nozzles that are attached to its weeding implement. The prototype described in [21] is designed to work with an image processing algorithm which allows the detection of weeds around finger millet and subsequent selective application of herbicide. The authors reported successful weed identification in almost all cases and that the duration of identification and spraying is approximately 3 s. In a similar study, a prototype robot for detecting weeds in onion fields is presented in [22]. Their proposed vision-based weed identification system successfully identified weeds in the onion field with approximately 97% accuracy. Another machine learning-based weed detection method is described in [23]. In this work, a four-wheeled robotic platform moves in a straight line in cotton field rows and uses its multiple spraying nozzles to deliver the herbicide at the recognised weed locations. It is reported that their method has a similar precision (over 97%) in identifying weeds. The Asterix robot’s targeted spraying has been tested in carrot fields [24]. The robot uses a forward-facing camera to detect the seed lines and apply a controlled quantity of herbicides when weeds are detected.
Taking into account the weed location provided by the machine vision methods described above, the spraying devices employed by these robots are designed to apply the chemical substance selectively. This results in both reducing the amount of chemical used and minimising its effect on the actual crop. For example, the two-wheeled intra-row robot described in [25] uses a novel targeted tool which involves the targeted application of herbicide after first damaging the weed tissue. With this configuration, the robot is able to direct herbicide delivery to working zones detected by the vision model. Experiments in a maise field revealed that the robot achieved a weed removal rate of up to 90%, a crop damage rate of less than 1.95% at a good working speed in a maise field, and a 94.45% weed removal rate and a 0.82% crop damage rate in a Chinese cabbage field. A combination of spraying and mechanical weeding can also be found in [26], where vision guides the spraying and stamping tool. Another selective spraying device is described in [27], where the simulated BonnBot-I robot is equipped with spot-spray nozzles to treat the detected weeds individually. Table 1 summarises the characteristics of spraying weeding robots. In all cases, vision was used for weed and crop detection.
Figure 2 shows examples of spraying weeding robots and an example of an implement for targeted spraying.

3.1.2. Non-Chemical Weeding Robots

As in the previous section, weeding robots in this category display various navigation methods, mainly dictated by the available sensing equipment and the target crop. The Korean K-Weedbot is able to extract morphological features (seedling characteristic points) from its vision module in order to determine the rows of paddy fields that will determine its navigation [28,29]. The authors of this study reported high precision guidance with a less than 1-degree error in the estimated guidance line. A similar navigational strategy is demonstrated for paddy fields by [30]. In recent work, a machine vision-based navigation method for a weeding robot is presented in [31]. In this case, vision is used for rice seedling detection, which leads to seedling line extraction with the least squares method and provides visual feedback for navigation corrections. Field experiments using the proposed method yielded an average weed control rate of 82.4% and a seedling injury rate of 2.8%. On the other hand, Cowbot, an autonomous weed mowing robot for maintaining cow pastures, uses algorithms for online coverage planning, which takes into account continuously retrieved information regarding weed detection in order to optimise path length and ensure coverage. The robot utilises sensor fusion from Real-Time Kinematic Positioning (RTK), an Inertial Measurement Unit (IMU) and two cameras to assist navigation [32]. Another sensor fusion method for the navigation of a weeding robot is presented in [33]. The proposed method attempts to integrate satellite-based location information, compass and machine vision to accurately guide the robot along a pre-defined route to cover the entire paddy field. Their algorithm was able to identify the paddy field rows and guide the robot with good accuracy (less than 2.5° in orientation error). In the case of [34], however, the authors selected to assign the navigation part to a human operator controlling the vehicle remotely, while the weeding operation is carried out autonomously with a camera locating the weeds and guiding a gripper for weed removal. In [35], the target environment is a cucumber greenhouse, so the weeding robot moves on an installed monorail. In [36], the proposed robot possesses a novel screw-type wheel design and wheel angle adjustment that provide better in-row navigation while weeding slurry paddy fields. Table 2 summarises these works.
In terms of sensing, it can be seen that weed detection using vision is predominant in the literature. A complete weeding robotic system is presented in [37]. In this work, a robot is able to navigate a cotton field by following waypoints, performing vision-based weed recognition and applying targeted treatment using both spraying and mechanical tools. The robot demonstrates an accuracy of more than 92% in recognising a range of weed species. In [38], the AgBotII robot is used for weed scouting. The objective of the proposed method is to enable the robot to be deployed in a field without prior knowledge of the target crop. Using collected data from field trials, the method uses machine vision approaches to cluster plants into groups using clustering algorithms. A planning algorithm for treating weeds using laser beams is demonstrated in [39]. While moving, the robot recognises the weeds within an area using vision, and the laser beam is directed to consecutive locations of detected weeds. The system displayed good accuracy with a weeding hit rate of up to 97%. In contrast with contact weed removal methods, for contactless laser weeding, an arm with only two degrees of freedom is sufficient. However, the accuracy and speed of the gimbals are important to ensure the accurate application of the beam. In this study, two Class 1 laser pointers were used to simulate the more powerful Class 4 lasers that would be used in weeding. Nevertheless, the proposed system achieved a moving speed of 30 mm per second while applying the beam for 0.64 s per weed. In [40], a vision-based system using the BoniRob platform is presented. The system uses two cameras: one camera in front of the manipulator to detect the weed and a second camera attached to the end effector for visual servoing to drive the weeding tool. The system was able to remove 1.75 weeds per second. The robotic system described in [41] exhibited good performance in recognising corn seedlings (up to 93% recognition rate) and weeds (up to 89% recognition rate) in corn fields, and this resulted in high weed prevention effectiveness and low seedling injury rate, given a laser beam as a weeding tool. In this case, the blue laser device is the end effector of a 5-degree-of-freedom arm, moving according to the robot’s moving platform and adjusting according to the detected weeds. The authors calculated the appropriate laser emission doses to apply in order to inhibit weed growth. The recognition accuracy was dependent upon the speed of the robot. The proposed robotic platform AGRIBOT is designed to perform autonomous navigation in a field for real-time weed detection [42]. In the relevant simulation work, the authors showed that their trained model successfully performed weed recognition with good performance in terms of accuracy (approximately 99%) and latency (2.5 frames per second), suitable for real-time weed detection. In a similar manner, the weed-detecting three-legged robot presented in [43] was able to identify weeds using a trained vision model with an identification rate above 99.5% in order to guide a delta arm for weed removal. Preliminary investigations on robots with machine vision for weed and crop discrimination can also be found in [44,45]. For weed detection, these studies use a Convolution Neural Network (CNN) and a fuzzy real-time classifier, respectively. Table 3 summarises the research described above.
Other studies have focused on the weeding tools employed by the weeding robots. For example, a weed suppression mechanism is presented in [46]. The robot possesses an arm with a brush at its end-effector, which applies force to weed seedlings, thus suppressing their growth. The authors have also considered controlling the posture of the robot when operating on uneven ground. Another early work describes the modular BoniRob platform within the context of the RemoteFarming project, which can accept various tools depending on the target crops [47]. The study presented in [48] compares three mechanical weeding implements mounted on an AgBotII robot, namely the arrow hoe, a tine, and a cutting tool, for treating cotton and grasses. The study found that, of the three tools examined, when treating cotton is considered, the most effective tools were the tine and the cutting mechanism, but the cutting mechanism was ineffective for grasses. Also, the experiments suggested that early treatment (four weeks after planting) is the most effective strategy. An approach using a floating robot frame in a paddy field to perform weeding is employed in [49]. In this approach, the teleoperated robot uses propellers for steering and maintaining its heading, while it uses chains to stir the soil under the water and thus dislocate the weeds. Using a similar principle, the paddy field weeding robot presented in [50] possesses specially designed wheels, which are the actual implements that are used to stir the soil and remove the weeds. In this case, however, the robot is moving autonomously using coverage planning based on the rice seedlings detected by capacitive sensors. This is in contrast to the vision-based approaches followed by the vast majority of the other approaches discussed in this paper. A robot moving on a conveyor belt using its rotating weeding tool (two vertically rotating discs with weeding knives) when weeds are detected is described in [51]. Using this tool, the authors report a weed removal rate of up to 84.4% in field trials, while they achieved a crop injury rate of around 1%. A prototype robot with a gesture-controlled weeding arm is proposed in [52]. Here, the weeding arm is taught to perform the weeding action by movements made by a hand glove. Table 4 summarises these aforementioned results.
The reviewed literature also includes works focusing on subjects that cannot be grouped with the above, such as performance comparisons. For example, a comparison between a commercial robotic lawnmower and other non-automated weed removal methods in terms of cost and effectiveness for pear orchards is presented in [53]. Specific measures have been employed to assess both the effectiveness and efficiency of the machines, including weed-cutting efficiency (ratio of cut weeds to the total weeds present) and total annual costs (including ownership, maintenance, energy, etc.). The authors found that for smaller fields, the robotic device is more cost-effective than other conventional lawnmowers, and it has displayed good performance in all field sizes. In a recent study, [54] the performance of seven robotic systems for weeding in sugar beet fields and winter oil-seed rape, compared to traditional herbicide treatment, was examined. Field experiments measured weed and crop density and working rate. The study found that robots reduced weed density by at least equal to the standard herbicide treatment. Also, in some cases, robotic treatment resulted in significant herbicide savings. Furthermore, it was found that in some experiments, robotic weeding achieved a weeding control efficacy of 93%, compared to the average of 83% for herbicide applications. The study concluded that despite the high cost of weeding robots, this approach is reliable and effective.
There have also been investigations on the environmental impacts of weeding robots. In 2015, a study investigated the fuel consumption of robotic weeding tractors for various weeding approaches [55]. A model to measure fuel consumption was developed, and methods related to adjustments in gear and throttle positions to reduce consumption were proposed. In a more recent study, energy use, crop yields and emissions were studied by comparing several non-chemical weeding methods with electric and diesel fuel robotic tractors. It was found that intelligent robotic weed control methods are more efficient compared to conventional ones [56]. More specifically, the lowest total energy consumption was achieved with vision-based mechanical inter-row loosening. This weeding method resulted in lower emissions when a diesel-powered robot was used. Also, the use of weeding robots resulted in higher sugar beet root yield compared to applying inter-row mulching. In a subsequent study, an evaluation of a conventional weeding system and FD20 robot was performed in terms of performance and effect on soil [57]. The results showed that the average weed control effectiveness inter-row was higher for the robotic system and that the robotic system had a minimal effect on soil penetration resistance, while the conventional weed control systems increased soil penetration resistance by up to 20%. In [58], a life cycle assessment (LCA) of the WeLASER weeding robot is presented in order to identify its strengths and weaknesses in environmental terms. It was found that even though autonomous laser-based weeding robots show potential for environmental efficiency, their energy issues are still the most challenging. Energy-related environmental impacts are found to be related mainly to thermal energy generation by the diesel engine machine, but the laser-weeding method has only a moderate environmental impact compared to mechanical and chemical weed control methods, as long as the method is used in a targeted manner. However, the study does not examine additional expected benefits of laser-based weeding, such as higher food quality and lower soil compaction. Table 5 summarises the approaches presented above.
Figure 3 shows examples of weeding robots with mechanical weeding tools.

3.1.3. Cooperative Approaches

Cooperative approaches, mainly multi-robot, have also been proposed in the literature, aimed at improving the efficiency and performance of the weeding systems. For example, in [59], the authors propose the use of the BoniRob robotic platform [60] as the Unmanned Ground Vehicle (UGV) with an Unmanned Aerial Vehicle (UAV). In this configuration, the UAV is used to identify weed pressure in the crops from the air and then communicate this information to the ground robot, which, in turn, moves towards the desired areas for targeted weed removal. To achieve this, a map that can be shared between the UAV and the UGV is constructed. To achieve this, the authors use a pipeline that registers heterogeneous georeferenced point clouds generated by heterogeneous vehicles. In practical terms, this approach aims at efficient weed removal interventions, where the UGV only visits the locations of interest, taking advantage of the more efficient UAV monitoring capabilities. In [61], a fleet of heterogeneous ground and aerial robots is employed for pest control. In this study, various weeding approaches are evaluated for different crops, including chemical spraying, mechanical weeding as well as thermal weeding. The fleet is coordinated by a mission manager, who includes planners for the aerial and ground teams. The mission manager can orchestrate both inspections and treatment missions according to the user’s mission parameters. It is responsible for generating robot trajectories, obtaining data from the perception systems and supervising the behaviour of the robots during the mission. Another multi-robot approach is proposed in the simulation work described in [62,63], where the focus is to develop, through a novel simulation software, an appropriate coordination planner. This work involves a centralised planner assigning tasks to identical agents who are recognising and killing weeds in the crop rows while navigating. The planner optimises performance by utilising information about the environment shared by the weeding agents so that the agents are directed to specific rows in the field based on a reward model. The simulation environment developed for this work was used to perform various computational experiments in a simulated field in order to investigate how parameters such as initial weed densities, weed growth duration, robot team size and time of deployment affect the effectiveness of the coordinated weeding. These parameters help design more effective multi-robot weeding interventions. Finally, a co-robot weeding approach is proposed in [64]. In this system, there is a human operator on the machine, but the operation of a pair of intra-row hoes is controlled automatically, according to the known plant positions. With accurate positioning, it is shown that it is possible to perform weeding while at the same time protecting the crop, and the system was shown to significantly reduce manual labour. The time required by the proposed system to hoe a specific intra-row region was measured to be 10.2 h per hectare, compared to the 24.1 h per hectare required by manual hoeing, indicating a significant time gain. Table 6 below summarises the reviewed cooperative approaches.

3.2. Examples of Commercial Weeding Platforms

In this section, robots developed by companies in the agricultural automation industry are examined. As in the previous section, two main approaches for weed control can be identified: chemical spraying and mechanical tools.
Chemical weed prevention has focused on optimising the amount of chemicals used by focusing on spraying in specific regions and thus controlling the quantity of chemicals used. For example, the company Ecorobotix is developing the solar-powered AVO robotic weeder, which uses targeted spraying for weeding [65]. According to the manufacturer, AVO can treat up to ten hectares a day by using up to 95% less weedkillers. The company EarthSense offers a range of robotic solutions for various tasks such as planting and phenotyping, and more specifically, weeding with their TerraMax platform [66]. Based on data retrieved from the website, the robot can achieve over 20% reduction in fertiliser costs. The robot uses proprietary computer vision for navigation, thus eliminating the need for RTK-GPS equipment for localisation.
In terms of commercial non-spraying robots, Naïo Technologies offers a range of weeding robots suitable for autonomous weeding, specialised for vineyards [67], with a selection of various weeding tools. The fully electrical Ted robot is reported to be able to treat up to 5 hectares per day, with 8 h of energy autonomy. The company Farming Revolution from Germany has developed the robot Farming GT [68], which is capable of weeding a wide variety of crops. Farming GT is fully autonomous, can be supervised remotely, and can treat up to 70 ha a week. The robot Maverick by odd.bot in the Netherlands is capable of autonomously removing weeds with mechanical weeding tools [69]. According to the manufacturer, the autonomous robot can remove up to two weeds per second per weeding arm, thus clearing one hectare in up to 16 h. Robotti, developed by Agrointelli in Italy, is a general-purpose autonomous fuel-based robot which is customised depending on the required task by installing the appropriate implement, with hoeing mechanisms available for weeding operations [70]. The commercial robotic weeder Tertil has been evaluated in [71]. The robot is suitable for small gardens and possesses capacitive sensors for obstacle avoidance, as well as weed whackers as a weeding tool. This robot was evaluated in experiments with two surrogate weed species, and weeding efficacy (taking into account pre-treatment and post-treatment weed densities) was measured. The robot, using its grousers and weed whacker, was able to achieve efficacies of up to 75%. Finally, there is the LaserWeeder robot by Carbon Robotics, which also utilises machine vision for weed detection and implements laser-based weed elimination for precision weeding [72]. It possesses 8 independent weeding modules, which are ready to fire every 50 ms. Currently, its fully autonomous version, including autonomous navigation, is available for demonstration purposes only. FD20 by FarmDroid in Denmark [73] is a solar-powered seeding and weeding robot that relies on RTK technology for navigation and mechanical add-on tools for actual weeding operations. With its solar panels, the robot can continuously operate for up to 24 h and can treat up to 6.5 hectares per day. Figure 4 shows representative examples of commercially available weeding robots.

3.3. Trends

In this section, a general overview of the aforementioned works is presented. The objective is to identify geographical and chronological trends in the field of weeding robots. In this analysis, only papers that report progress on some specific system or method are included, and therefore, review papers are omitted.
Initially, the geographical distribution of the papers reviewed here was examined. The affiliation of the principal author was used to determine the country of origin of a publication with authors from multiple countries. Figure 5 shows a chart listing the number of research papers and commercial robots per country. The graph is used to highlight the relationship between a country’s agricultural needs and the corresponding research and commercial output in weeding robots.
Figure 6 illustrates the distribution of works examined in this paper per geographical region.
Figure 7 shows the number of papers related to weeding robots published in each year since 2014. In this graph, the commercial platforms are not included since they are still available after their initial release. Note that none of the papers reviewed were published in 2016.

4. Discussion

This paper reviewed the research efforts carried out in the field of weeding robotics during the last ten years. The platforms developed were presented, and their proposed solutions in the specific areas of locomotion and navigation, sensing and weeding tools were discussed. In addition, examples of currently available commercial robotic platforms were also given.
The review has revealed that much of the research is gradually moving away from the use of chemical weeding. Most of the research works have focused on mechanical or laser-based tools. This trend can be explained by the rising herbicide costs, the various health and environmental concerns and the rise of herbicide-resistant crops. Even though only one of the reviewed papers [58] directly compares the effectiveness of chemical and non-chemical weed removal methods, non-chemical robots demonstrate high rates of weed removal without the negative effects of chemicals while avoiding the cost of herbicides. This reinforces the notion that non-chemical weeding is a promising alternative. Although laser devices as the robot’s weeding implement are explored only in a few of the studies reviewed here, optical (laser-based) weed control begins to appear more frequently in the more recent studies. As vision-based weed detection is becoming more effective and more precise, laser-based systems’ high precision localisation capabilities and high speeds of treatment can be exploited [74]. A study presented in [75] has shown that European farmers have shown interest in adopting such technologies given their efficiency, precision and positive impact on food safety.
In almost all cases, computer vision was used for weed and crop identification in order to apply weed control in a more thorough and targeted manner. Two main approaches have been identified: (a) using traditional image processing algorithms (feature-based, classification, classifiers, etc.) and (b) data-driven deep learning methods. The use of targeted approaches, both chemical and non-chemical, is supported and aided by advances in machine vision. Newer and more effective machine learning algorithms, widely available training datasets and increasingly computationally capable onboard computers allow recognition, classification and localisation of weeds and crops to be performed by mobile robotic platforms in real-time, guiding the corresponding weeding tools without compromising the operating speed. Deep learning approaches, in particular, attempt to improve their performance by investigating the various neural network structures and their parameters, the training strategies and creating more complete datasets [76]. Especially in the case of mechanical and laser-based weeding, precise administration of weed control is increasingly possible. Such computer vision methods are expected to become the standard approach in robotic weeding, given the amount of research conducted in this sub-field of artificial intelligence, the promising results so far, and the improvements in effectiveness. Through those means, weeding robots will be able to distinguish between crops and weeds and distinguish between weed species in various stages of development [77]. The weeding tools themselves are evolving in accordance with the precision weed treatment capabilities provided by machine vision. This is especially the case with weeding tools attached, as end effectors, to robotic arms with enough degrees of freedom to guide the tool to precise locations. This is further reinforced by novel targeted treatment approaches, such as laser-based weeding tools, as discussed above.
In terms of autonomous navigation, the navigational strategies employed depend on the target crop. In general, weeding robots are observed to employ two main types of sensors that aid navigation: (a) cameras and (b) RTK-GPS systems. In the former case, research is focused on identifying the surroundings of the robot (most commonly to identify a crop row by recognising the seedlings) and thus dictate the direction in which the robot should move. In the latter case, using established localisation technologies, previous knowledge of a static crop field is required so that the robot can follow a prescribed path, determine its previously visited positions when performing an exhaustive search of the environment, or determine the precise location of the weeds or seedlings. For applications which require increased accuracy in localisation, the cost of the required RTK-GPS system also increases significantly. Where required, other sensors (bumpers, laser scanners, capacitive sensors, etc.) have also been employed to aid in local obstacle avoidance and ensure the safety of both the robots and the crop.
A number of the studies reviewed in this paper have utilised general-purpose machines such as the BoniRob [60] or the Thorvald [78] robots, which provide autonomous capabilities but can be adjusted with a variety of implements depending on the target crop or the desired agricultural task. Other studies have opted to develop custom robotic platforms that have been designed to meet the desired specifications. The literature review reveals that most robots operate in typical orchards and that crops are arranged in rows. These are well-defined and facilitate the application of vision-based row detection algorithms to perform navigation. For this reason, the robots employed in this case are predominantly wheeled with three or four wheels of appropriate size to perform in-row navigation. The width of the robots depends on the spatial arrangement of the target crop, so not all robots are suitable for all crop row arrangements. This is a challenge that needs to be overcome before weeding robots can become more widespread in agricultural production. Modular robots can aid in this direction by accommodating weeding implements of adjustable size. Crop rows are also most commonly found in plots of land that are typically flat, without large inclinations of the surface. Only one study [46] has examined the existence of soil irregularities and proposed a method to remedy its negative effects on weed implement positioning. There was also a study in which a robotic lawnmower operated on steep slopes in a pasture [32]. The authors, in this case, have adopted a practical solution by adapting a proven commercial lawn mower to perform autonomous coverage of the entire area of the pasture. Innovative robot designs have been proposed for very specific operating environments, such as the floating robotic frame for flooded rice paddies [48].
An important consideration in developing robots and ensuring their autonomy is to fulfil the power requirements for weeding tasks. Such tasks are usually labour-intensive and time-consuming over large areas. Robots reviewed in this paper have been powered either by electric or by fuel motors. While fuel-based machines have traditionally been used in agriculture and fulfil the requirements for long duration of operation and adequate power output, current electric motors display good performance. Unless specialised weeding implements or other onboard equipment that require large amounts of power are used, batteries are the preferred source of power for robots because of their minimal environmental impact and the elimination of fuel costs. Additionally, the use of solar panels further augments the robot’s autonomy. In one of the studies, a strategy for automatically recharging a low-cost weeding robot has been proposed [18].
Regarding the observed trends, it can be seen that most studies originate in countries like the USA, China and India. These are countries with strong agricultural sectors, so it follows that they have an equally strong motivation to pursue the automation of all agricultural operations and weeding operations in particular. This relationship between region and agricultural research is further reinforced by examining the types of crops which are the focus of research per country. For example, in countries in Asia such as China or Japan, which traditionally produce rice, most of the research is targeted to weeds in paddy fields. On the other hand, research conducted in countries such as the USA and Germany shows that other crops are favoured, such as beets and carrots. As a result, crop-specific technologies are being developed in each country. For example, in countries where the crop is arranged in rows, research on appropriate algorithms for crop navigation is more prevalent. Furthermore, an examination of the number of papers produced per year since 2014 reveals that the interest in the development of weeding robots is not diminishing. This means that a) there is an interest in pursuing this line of research based on the needs of the industry, and b) there are still challenges that need to be addressed in the field.

5. Conclusions

The task of weeding is challenging and complex, and this is especially true when robotic weeding is considered, given current technological limitations [79]. This is confirmed by the reviewed research, even though it is obvious that significant progress has been made in this field. This progress is also demonstrated by the availability of reliable commercial weeding robots.
However, based on the reviewed works, three main challenges that need to be addressed have been identified. Firstly, the targeted application of weed control, either chemical or mechanical, is key for implementing effective and efficient weeding robots. Machine vision is a major field of active research that aims to produce more effective models for successful weed and crop recognition in order to increase weed removal rates and reduce damage to crops. Environmental conditions need to be taken into account in order to produce robust and reliable models, and this will depend on (a) better quality of current datasets, (b) availability of datasets for a wide variety of weed types and crops, and (c) further improvements on the current training methods. Advances in artificial intelligence and machine vision models, as well as vision sensors, are expected to further improve the performance of vision-based systems for both weed and crop identification and vision-based navigation. Secondly, the standardisation of weeding tools and weeding procedures can take advantage of existing proven platforms, which can be adjusted for weeding applications. These modular platforms can take advantage of weeding tools, which will be used interchangeably depending on the target crop, together with the corresponding pre-trained vision model, as required. Thirdly, full autonomy is the ultimate goal of engaging robots in agriculture. This will require autonomous operation in terms of navigation within the field and autonomous weeding in terms of energy autonomy. These issues can be addressed by improved localisation and navigation algorithms, more accurate and effective detection methods and removal tools, and new, more efficient battery technologies and autonomous recharging techniques.

Author Contributions

Conceptualization, C.L. and T.P.; methodology, C.L. and T.P.; validation, C.L.; formal analysis, C.L; writing—original draft preparation, C.L.; writing—review and editing, C.L. and T.P.; supervision, T.P. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. United Nations World Population Projected to Reach 9.8 Billion in 2050. Available online: https://www.un.org/development/desa/en/news/population/world-population-prospects-2017.html (accessed on 12 December 2022).
  2. van Dijk, M.; Morley, T.; Rau, M.L.; Saghai, Y. A Meta-Analysis of Projected Global Food Demand and Population at Risk of Hunger for the Period 2010–2050. Nat. Food 2021, 2, 494–501. [Google Scholar] [CrossRef] [PubMed]
  3. Droukas, L.; Doulgeri, Z.; Tsakiridis, N.L.; Triantafyllou, D.; Kleitsiotis, I.; Mariolis, I.; Giakoumis, D.; Tzovaras, D.; Kateris, D.; Bochtis, D. A Survey of Robotic Harvesting Systems and Enabling Technologies. J. Intell. Robot. Syst. 2023, 107, 21. [Google Scholar] [CrossRef] [PubMed]
  4. Oliveira, L.F.P.; Moreira, A.P.; Silva, M.F. Advances in Agriculture Robotics: A State-of-the-Art Review and Challenges Ahead. Robotics 2021, 10, 52. [Google Scholar] [CrossRef]
  5. Bechar, A. (Ed.) Innovation in Agricultural Robotics for Precision Agriculture; Progress in Precision Agriculture; Springer International Publishing: Cham, Switzerland, 2021; ISBN 978-3-030-77035-8. [Google Scholar]
  6. dos Santos, F.N.; Sobreira, H.; Campos, D.; Morais, R.; Paulo Moreira, A.; Contente, O. Towards a Reliable Robot for Steep Slope Vineyards Monitoring. J. Intell. Robot. Syst. 2016, 83, 429–444. [Google Scholar] [CrossRef]
  7. Reiser, D.; Vázquez-Arellano, M.; Paraforos, D.S.; Garrido-Izard, M.; Griepentrog, H.W. Iterative Individual Plant Clustering in Maize with Assembled 2D LiDAR Data. Comput. Ind. 2018, 99, 42–52. [Google Scholar] [CrossRef]
  8. Xiong, Y.; Ge, Y.; Grimstad, L.; From, P.J. An Autonomous Strawberry-harvesting Robot: Design, Development, Integration, and Field Evaluation. J. Field Robot. 2020, 37, 202–224. [Google Scholar] [CrossRef]
  9. Starostin, I.A.; Eshchin, A.V.; Davydova, S.A. Global Trends in the Development of Agricultural Robotics. IOP Conf. Ser. Earth Environ. Sci. 2023, 1138, 012042. [Google Scholar] [CrossRef]
  10. Radoglou-Grammatikis, P.; Sarigiannidis, P.; Lagkas, T.; Moscholios, I. A Compilation of UAV Applications for Precision Agriculture. Comput. Netw. 2020, 172, 107148. [Google Scholar] [CrossRef]
  11. Lytridis, C.; Kaburlasos, V.G.; Pachidis, T.; Manios, M.; Vrochidou, E.; Kalampokas, T.; Chatzistamatis, S. An Overview of Cooperative Robotics in Agriculture. Agronomy 2021, 11, 1818. [Google Scholar] [CrossRef]
  12. Mao, W.; Liu, Z.; Liu, H.; Yang, F.; Wang, M. Research Progress on Synergistic Technologies of Agricultural Multi-Robots. Appl. Sci. 2021, 11, 1448. [Google Scholar] [CrossRef]
  13. Fennimore, S.A.; Cutulle, M. Robotic Weeders Can Improve Weed Control Options for Specialty Crops. Pest Manag. Sci. 2019, 75, 1767–1774. [Google Scholar] [CrossRef] [PubMed]
  14. Gerhards, R.; Andújar Sanchez, D.; Hamouz, P.; Peteinatos, G.G.; Christensen, S.; Fernandez-Quintanilla, C. Advances in Site-specific Weed Management in Agriculture—A Review. Weed Res. 2022, 62, 123–133. [Google Scholar] [CrossRef]
  15. Li, N.; Zhang, X.; Zhang, C.; Ge, L.; He, Y.; Wu, X. Review of Machine-Vision-Based Plant Detection Technologies for Robotic Weeding. In Proceedings of the 2019 IEEE International Conference on Robotics and Biomimetics (ROBIO), Dali, China, 6–8 December 2019; pp. 2370–2377. [Google Scholar] [CrossRef]
  16. Zhang, W.; Miao, Z.; Li, N.; He, C.; Sun, T. Review of Current Robotic Approaches for Precision Weed Management. Curr. Robot. Rep. 2022, 3, 139–151. [Google Scholar] [CrossRef] [PubMed]
  17. Li, Y.; Guo, Z.; Shuang, F.; Zhang, M.; Li, X. Key Technologies of Machine Vision for Weeding Robots: A Review and Benchmark. Comput. Electron. Agric. 2022, 196, 106880. [Google Scholar] [CrossRef]
  18. Du, Y.; Mallajosyula, B.; Sun, D.; Chen, J.; Zhao, Z.; Rahman, M.; Quadir, M.; Jawed, M.K. A Low-Cost Robot with Autonomous Recharge and Navigation for Weed Control in Fields with Narrow Row Spacing. In Proceedings of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 27 September–1 October 2021; IEEE: Piscataway, NJ, USA; pp. 3263–3270. [Google Scholar]
  19. Sabanci, K.; Aydin, C. Smart Robotic Weed Control System for Sugar Beet. J. Agric. Sci. Technol. 2017, 19, 73–83. [Google Scholar]
  20. Scholz, C.; Kohlbrecher, M.; Ruckelshausen, A.; Kinski, D.; Mentrup, D. Camera-Based Selective Weed Control Application Module (“Precision Spraying App”) for the Autonomous Field Robot Platform BoniRob. In Proceedings of the International Conference of Agricultural Engineering, Zurich, Switzerand, 6–10 July 2014; pp. 1–8. [Google Scholar]
  21. Aravind, R.; Daman, M.; Kariyappa, B.S. Design and Development of Automatic Weed Detection and Smart Herbicide Sprayer Robot. In Proceedings of the 2015 IEEE Recent Advances in Intelligent Computational Systems (RAICS), Trivandrum, Kerala, India, 10–12 December 2015; IEEE: Piscataway, NJ, USA; pp. 257–261. [Google Scholar]
  22. Arakeri, M.P.; Vijaya Kumar, B.P.; Barsaiya, S.; Sairam, H.V. Computer Vision Based Robotic Weed Control System for Precision Agriculture. In Proceedings of the 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Manipal, Karnataka, India, 13–16 September 2017; IEEE: Piscataway, NJ, USA; pp. 1201–1205. [Google Scholar]
  23. Fan, X.; Chai, X.; Zhou, J.; Sun, T. Deep Learning Based Weed Detection and Target Spraying Robot System at Seedling Stage of Cotton Field. Comput. Electron. Agric. 2023, 214, 108317. [Google Scholar] [CrossRef]
  24. Utstumo, T.; Urdal, F.; Brevik, A.; Dørum, J.; Netland, J.; Overskeid, Ø.; Berge, T.W.; Gravdahl, J.T. Robotic In-Row Weed Control in Vegetables. Comput. Electron. Agric. 2018, 154, 36–45. [Google Scholar] [CrossRef]
  25. Jiang, W.; Quan, L.; Wei, G.; Chang, C.; Geng, T. A Conceptual Evaluation of a Weed Control Method with Post-Damage Application of Herbicides: A Composite Intelligent Intra-Row Weeding Robot. Soil Tillage Res. 2023, 234, 105837. [Google Scholar] [CrossRef]
  26. Wu, X.; Aravecchia, S.; Lottes, P.; Stachniss, C.; Pradalier, C. Robotic Weed Control Using Automated Weed and Crop Classification. J. Field Robot. 2020, 37, 322–340. [Google Scholar] [CrossRef]
  27. Ahmadi, A.; Halstead, M.; McCool, C. BonnBot-I: A Precise Weed Management and Crop Monitoring Platform. In Proceedings of the 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Kyoto, Japan, 23–27 October 2022; IEEE: Piscataway, NJ, USA; pp. 9202–9209. [Google Scholar]
  28. Choi, K.H.; Han, S.K.; Han, S.H.; Park, K.-H.; Kim, K.-S.; Kim, S. Morphology-Based Guidance Line Extraction for an Autonomous Weeding Robot in Paddy Fields. Comput. Electron. Agric. 2015, 113, 266–274. [Google Scholar] [CrossRef]
  29. Choi, K.H.; Han, S.K.; Park, K.-H.; Kim, K.-S.; Kim, S. Vision Based Guidance Line Extraction for Autonomous Weed Control Robot in Paddy Field. In Proceedings of the 2015 IEEE International Conference on Robotics and Biomimetics (ROBIO), Zhuhai China, 6–9 December 2015; IEEE: Piscataway, NJ, USA; pp. 831–836. [Google Scholar]
  30. Zhang, Q.; Shaojie Chen, M.E.; Li, B. A Visual Navigation Algorithm for Paddy Field Weeding Robot Based on Image Understanding. Comput. Electron. Agric. 2017, 143, 66–78. [Google Scholar] [CrossRef]
  31. Ju, J.; Chen, G.; Lv, Z.; Zhao, M.; Sun, L.; Wang, Z.; Wang, J. Design and Experiment of an Adaptive Cruise Weeding Robot for Paddy Fields Based on Improved YOLOv5. Comput. Electron. Agric. 2024, 219, 108824. [Google Scholar] [CrossRef]
  32. Maini, P.; Gonultas, B.M.; Isler, V. Online Coverage Planning for an Autonomous Weed Mowing Robot With Curvature Constraints. IEEE Robot. Autom. Lett. 2022, 7, 5445–5452. [Google Scholar] [CrossRef]
  33. Kanagasingham, S.; Ekpanyapong, M.; Chaihan, R. Integrating Machine Vision-Based Row Guidance with GPS and Compass-Based Routing to Achieve Autonomous Navigation for a Rice Field Weeding Robot. Precis. Agric. 2020, 21, 831–855. [Google Scholar] [CrossRef]
  34. Visentin, F.; Cremasco, S.; Sozzi, M.; Signorini, L.; Signorini, M.; Marinello, F.; Muradore, R. A Mixed-Autonomous Robotic Platform for Intra-Row and Inter-Row Weed Removal for Precision Agriculture. Comput. Electron. Agric. 2023, 214, 108270. [Google Scholar] [CrossRef]
  35. Heravi, A.; Ahmad, D.; Hameed, I.A.; Shamshiri, R.R.; Balasundram, S.; Yamin, M. Development of a Field Robot Platform for Mechanical Weed Control in Greenhouse Cultivation of Cucumber. In Agricultural Robots: Fundamentals and Applications; IntechOpen: London, UK, 2019. [Google Scholar]
  36. Murugaraj, G.; Selva Kumar, S.; Pillai, A.S.; Bharatiraja, C. Implementation of In-Row Weeding Robot with Novel Wheel, Assembly and Wheel Angle Adjustment for Slurry Paddy Field. Mater. Today Proc. 2022, 65, 215–220. [Google Scholar] [CrossRef]
  37. Bawden, O.; Kulk, J.; Russell, R.; McCool, C.; English, A.; Dayoub, F.; Lehnert, C.; Perez, T. Robot for Weed Species Plant-specific Management. J. Field Robot. 2017, 34, 1179–1199. [Google Scholar] [CrossRef]
  38. Hall, D.; Dayoub, F.; Kulk, J.; McCool, C. Towards Unsupervised Weed Scouting for Agricultural Robotics. In Proceedings of the 2017 IEEE International Conference on Robotics and Automation (ICRA), Singapore, 29 May–3 June 2017; pp. 5223–5230. [Google Scholar] [CrossRef]
  39. Xiong, Y.; Ge, Y.; Liang, Y.; Blackmore, S. Development of a Prototype Robot and Fast Path-Planning Algorithm for Static Laser Weeding. Comput. Electron. Agric. 2017, 142, 494–503. [Google Scholar] [CrossRef]
  40. Michaels, A.; Haug, S.; Albert, A. Vision-Based High-Speed Manipulation for Robotic Ultra-Precise Weed Control. In Proceedings of the 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Hamburg, Germany, 28 September–2 October 2015; IEEE: Piscataway, NJ, USA; pp. 5498–5505. [Google Scholar]
  41. Zhu, H.; Zhang, Y.; Mu, D.; Bai, L.; Zhuang, H.; Li, H. YOLOX-Based Blue Laser Weeding Robot in Corn Field. Front. Plant Sci. 2022, 13, 1017803. [Google Scholar] [CrossRef]
  42. Patel, D.; Gandhi, M.; Shankaranarayanan, H.; Darji, A.D. Design of an Autonomous Agriculture Robot for Real-Time Weed Detection Using CNN. In Advances in VLSI and Embedded Systems; Lecture Notes in Electrical Engineering; Springer: Berlin/Heidelberg, Germany, 2023; Volume 962, pp. 141–161. ISBN 9789811967795. [Google Scholar]
  43. Sethia, G.; Guragol, H.K.S.; Sandhya, S.; Shruthi, J.; Rashmi, N. Automated Computer Vision Based Weed Removal Bot. In Proceedings of the 2020 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT), Bangalore, India, 2–4 July 2020; IEEE: Piscataway, NJ, USA; pp. 1–6. [Google Scholar]
  44. Florance Mary, M.; Yogaraman, D. Neural Network Based Weeding Robot For Crop And Weed Discrimination. J. Phys. Conf. Ser. 2021, 1979, 012027. [Google Scholar] [CrossRef]
  45. Sujaritha, M.; Annadurai, S.; Satheeshkumar, J.; Kowshik Sharan, S.; Mahesh, L. Weed Detecting Robot in Sugarcane Fields Using Fuzzy Real Time Classifier. Comput. Electron. Agric. 2017, 134, 160–171. [Google Scholar] [CrossRef]
  46. Nakai, S.; Yamada, Y. Development of a Weed Suppression Robot for Rice Cultivation: Weed Suppression and Posture Control. Int. J. Electr. Comput. Electron. Commun. Eng. 2014, 8, 1766–1770. [Google Scholar]
  47. Sellmann, F.; Bangert, W.; Grzonka, S.; Hänsel, M.; Haug, S.; Kielhorn, A.; Michaels, A.; Möller, K.; Rahe, F.; Strothmann, W.; et al. RemoteFarming.1: Human-Machine Interaction for a Field-Robot-Based Weed Control Application in Organic Farming. In Proceedings of the 4th International Conference on Machine Control & Guidance, Braunschweig, Germany, 19–20 March 2014. [Google Scholar]
  48. McCool, C.S.; Beattie, J.; Firn, J.; Lehnert, C.; Kulk, J.; Bawden, O.; Russell, R.; Perez, T. Efficacy of Mechanical Weeding Tools: A Study into Alternative Weed Management Strategies Enabled by Robotics. IEEE Robot. Autom. Lett. 2018, 3, 1184–1190. [Google Scholar] [CrossRef]
  49. Uchida, H.; Funaki, T.; Yamano, T. Development of a Remoto Control Type Weeding Machine with Stirring Chains for a Paddy Field. In Proceedings of the 22nd International Conference on Climbing and Walking Robots and the Support Technologies for Mobile Machines (CLAWAR 2019), Kuala Lumpur, Malaysia, 26–28 August 2019; CLAWAR Association Ltd.: Buckinghamshire, UK; pp. 26–28. [Google Scholar]
  50. Sori, H.; Inoue, H.; Hatta, H.; Ando, Y. Effect for a Paddy Weeding Robot in Wet Rice Culture. J. Robot. Mechatron. 2018, 30, 198–205. [Google Scholar] [CrossRef]
  51. Quan, L.; Jiang, W.; Li, H.; Li, H.; Wang, Q.; Chen, L. Intelligent Intra-Row Robotic Weeding System Combining Deep Learning Technology with a Targeted Weeding Mode. Biosyst. Eng. 2022, 216, 13–31. [Google Scholar] [CrossRef]
  52. Gokul, S.; Dhiksith, R.; Sundaresh, S.A.; Gopinath, M. Gesture Controlled Wireless Agricultural Weeding Robot. In Proceedings of the 2019 5th International Conference on Advanced Computing & Communication Systems (ICACCS), Coimbatore, India, 15–16 March 2019; IEEE: Piscataway, NJ, USA; pp. 926–929. [Google Scholar]
  53. Hossain, M.Z.; Komatsuzaki, M. Weed Management and Economic Analysis of a Robotic Lawnmower: A Case Study in a Japanese Pear Orchard. Agriculture 2021, 11, 113. [Google Scholar] [CrossRef]
  54. Gerhards, R.; Risser, P.; Spaeth, M.; Saile, M.; Peteinatos, G. A Comparison of Seven Innovative Robotic Weeding Systems and Reference Herbicide Strategies in Sugar Beet (Beta vulgaris Subsp. vulgaris L.) and Rapeseed (Brassica napus, L.). Weed Res. 2024, 64, 42–53. [Google Scholar] [CrossRef]
  55. Gonzalez-de-Soto, M.; Emmi, L.; Garcia, I.; Gonzalez-de-Santos, P. Reducing Fuel Consumption in Weed and Pest Control Using Robotic Tractors. Comput. Electron. Agric. 2015, 114, 96–113. [Google Scholar] [CrossRef]
  56. Bručienė, I.; Aleliūnas, D.; Šarauskis, E.; Romaneckas, K. Influence of Mechanical and Intelligent Robotic Weed Control Methods on Energy Efficiency and Environment in Organic Sugar Beet Production. Agriculture 2021, 11, 449. [Google Scholar] [CrossRef]
  57. Bručienė, I.; Buragienė, S.; Šarauskis, E. Weeding Effectiveness and Changes in Soil Physical Properties Using Inter-Row Hoeing and a Robot. Agronomy 2022, 12, 1514. [Google Scholar] [CrossRef]
  58. Krupanek, J.; de Santos, P.G.; Emmi, L.; Wollweber, M.; Sandmann, H.; Scholle, K.; Di Minh Tran, D.; Schouteten, J.J.; Andreasen, C. Environmental Performance of an Autonomous Laser Weeding Robot—A Case Study. Int. J. Life Cycle Assess. 2024, 29, 1021–1052. [Google Scholar] [CrossRef]
  59. Pretto, A.; Aravecchia, S.; Burgard, W.; Chebrolu, N.; Dornhege, C.; Falck, T.; Fleckenstein, F.; Fontenla, A.; Imperoli, M.; Khanna, R.; et al. Building an Aerial–Ground Robotics System for Precision Farming: An Adaptable Solution. IEEE Robot. Autom. Mag. 2021, 28, 29–49. [Google Scholar] [CrossRef]
  60. Ruckelshausen, A.; Biber, P.; Dorna, M.; Gremmes, H.; Klose, R.; Linz, A.; Rahe, R.; Resch, R.; Thiel, M.; Trautz, D.; et al. BoniRob: An Autonomous Field Robot Platform for Individual Plant Phenotyping. In Precision agriculture ’09; Brill|Wageningen Academic: Leiden, The Netherlands, 2009; pp. 841–847. ISBN 9789086861132. [Google Scholar]
  61. Gonzalez-de-Santos, P.; Ribeiro, A.; Fernandez-Quintanilla, C.; Lopez-Granados, F.; Brandstoetter, M.; Tomic, S.; Pedrazzi, S.; Peruzzi, A.; Pajares, G.; Kaplanis, G.; et al. Fleets of Robots for Environmentally-Safe Pest Control in Agriculture. Precis. Agric. 2017, 18, 574–614. [Google Scholar] [CrossRef]
  62. McAllister, W.; Osipychev, D.; Davis, A.; Chowdhary, G. Agbots: Weeding a Field with a Team of Autonomous Robots. Comput. Electron. Agric. 2019, 163, 104827. [Google Scholar] [CrossRef]
  63. McAllister, W.; Osipychev, D.; Chowdhary, G.; Davis, A. Multi-Agent Planning for Coordinated Robotic Weed Killing. In Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 1–5 October 2018; IEEE: Piscataway, NJ, USA; pp. 7955–7960. [Google Scholar]
  64. Pérez-Ruíz, M.; Slaughter, D.C.; Fathallah, F.A.; Gliever, C.J.; Miller, B.J. Co-Robotic Intra-Row Weed Control System. Biosyst. Eng. 2014, 126, 45–55. [Google Scholar] [CrossRef]
  65. Ecorobotix AVO. Available online: https://ecorobotix.com/en/avo/ (accessed on 9 June 2024).
  66. EarthSense TerraMax. Available online: https://www.earthsense.co (accessed on 9 June 2024).
  67. Naïo Technologies Ted. Available online: https://www.naio-technologies.com (accessed on 9 June 2024).
  68. Farming Revolution Farming GT. Available online: https://farming-revolution.com (accessed on 9 June 2024).
  69. odd.bot Maverick. Available online: https://www.odd.bot (accessed on 9 June 2024).
  70. Agrointelli Robotti. Available online: https://agrointelli.com/robotti/ (accessed on 9 June 2024).
  71. Sanchez, J.; Gallandt, E.R. Functionality and Efficacy of Franklin Robotics’ Tertill™ Robotic Weeder. Weed Technol. 2021, 35, 166–170. [Google Scholar] [CrossRef]
  72. Carbon Robotics. Autonomous LaserWeeder. Available online: https://carbonrobotics.com/autonomous-weeder (accessed on 9 June 2024).
  73. FarmDroid FD20. Available online: https://farmdroid.com (accessed on 9 June 2024).
  74. Zhang, H.; Cao, D.; Zhou, W.; Currie, K. Laser and Optical Radiation Weed Control: A Critical Review. Precis. Agric. 2024, 25, 2033–2057. [Google Scholar] [CrossRef]
  75. Tran, D.; Schouteten, J.J.; Degieter, M.; Krupanek, J.; Jarosz, W.; Areta, A.; Emmi, L.; De Steur, H.; Gellynck, X. European Stakeholders’ Perspectives on Implementation Potential of Precision Weed Control: The Case of Autonomous Vehicles with Laser Treatment. Precis. Agric. 2023, 24, 2200–2222. [Google Scholar] [CrossRef]
  76. Qu, H.-R.; Su, W.-H. Deep Learning-Based Weed–Crop Recognition for Smart Agricultural Equipment: A Review. Agronomy 2024, 14, 363. [Google Scholar] [CrossRef]
  77. Zingsheim, M.L.; Döring, T.F. What Weeding Robots Need to Know about Ecology. Agric. Ecosyst. Environ. 2024, 364, 108861. [Google Scholar] [CrossRef]
  78. Grimstad, L.; From, P. The Thorvald II Agricultural Robotic System. Robotics 2017, 6, 24. [Google Scholar] [CrossRef]
  79. Merfield, C.N. Robotic Weeding’s False Dawn? Ten Requirements for Fully Autonomous Mechanical Weed Management. Weed Res. 2016, 56, 340–344. [Google Scholar] [CrossRef]
Figure 1. Number of papers per type of publication.
Figure 1. Number of papers per type of publication.
Agriengineering 06 00187 g001
Figure 2. Examples of spraying weeding robots and spraying devices: (a) Targeted spraying in cotton fields [23], (b) The Asterix robot [24], (c) The weed control unit of the Flourish system [26].
Figure 2. Examples of spraying weeding robots and spraying devices: (a) Targeted spraying in cotton fields [23], (b) The Asterix robot [24], (c) The weed control unit of the Flourish system [26].
Agriengineering 06 00187 g002
Figure 3. Weeding robots using mechanical tools: (a) Intra-row robotic weeding [51], (b) The BoniRob platform using RemoteFarming.1 App [47].
Figure 3. Weeding robots using mechanical tools: (a) Intra-row robotic weeding [51], (b) The BoniRob platform using RemoteFarming.1 App [47].
Agriengineering 06 00187 g003
Figure 4. Examples of commercial weeding robots. (a) ecorobotics AVO [65]; (b) Farming GT [68]; (c) LaserWeeder [72].
Figure 4. Examples of commercial weeding robots. (a) ecorobotics AVO [65]; (b) Farming GT [68]; (c) LaserWeeder [72].
Agriengineering 06 00187 g004
Figure 5. Number of publications per country.
Figure 5. Number of publications per country.
Agriengineering 06 00187 g005
Figure 6. Allocation of publications per region.
Figure 6. Allocation of publications per region.
Agriengineering 06 00187 g006
Figure 7. Number of publications per year.
Figure 7. Number of publications per year.
Agriengineering 06 00187 g007
Table 1. Approaches for chemical-based weeding robots.
Table 1. Approaches for chemical-based weeding robots.
Type of StudyTarget CropWeed RecognitionWeeding ToolNavigationRef.
Field testingFlaxN/ASpraying systemVision for crop row detection[18]
Lab testingSugar beetsImage processingSpraying unitMoving on rail[19]
Lab and field testsN/AImage processingEight spraying nozzlesManual control[20]
Lab testingRagiImage processingSpraying systemN/A[21]
Lab testingOnionsImage processingSingle spraying toolRemote control[22]
Field testingCottonDeep learningSpraying nozzlesRow navigation[23]
Field testingCarrotsImage processingControlled sprayingVision for crop row detection[24]
Lab and field testsMaizeDeep learningA weeding tool with a sprayer and brushes to direct herbicide delivery to working zonesContinuously operating reference station (CORS)-based navigation system[25]
Lab testingSugar beetsDeep learningSelective sprayer and mechanical stamping toolN/A[26]
Lab and field testsN/ADeep learningSpot-spray nozzlesIn-row movements with plants at regular intervals[27]
Table 2. Summary of research regarding non-chemical weeding robots where navigation issues are investigated.
Table 2. Summary of research regarding non-chemical weeding robots where navigation issues are investigated.
Type of StudyTarget CropWeed RecognitionWeeding ToolNavigationRef.
Field testingPaddy fieldImage processingScrew-type wheels for weed removalVision for seedling line detection[28,29]
Lab testingPaddy fieldImage processingN/AVision for seedling line detection[30]
Field testingPaddy fieldDeep learningCultivator-weeding wheelsVision for seedling line detection[31]
Field testingCow pastureN/AFlail-deck weeding implementOnline coverage using RTK, IMU, and vision[32]
Field testingPaddy fieldImage processingThree-row paddy weeder with cutter blades(a) GNSS path planning, (b) compass bearing correction and (c) vision-based row guidance[33]
Lab and field testingVariety of plantsDeep learningGripper for weed pickingTeleoperation[34]
Field testingCucumberN/ARotating cutting bladeMonorail[35]
System overviewPaddy fieldN/AN/AN/A[36]
Table 3. Research of non-chemical weeding robots where sensing is the main focus.
Table 3. Research of non-chemical weeding robots where sensing is the main focus.
Type of StudyTarget CropWeed RecognitionWeeding ToolNavigationRef.
Field testingCottonImage processing1 DOF and 2 DOF weeding mechanismsCoverage planner[37]
Lab testingCottonDeep learningN/AN/A[38]
Field testingCloverImage processing2 DOFs arm with dual-gimbal laser pointers areaThe robot is weeding while static within a predefined frame captured by a camera[39]
Field testingN/AImage processingFlywheel stamping toolN/A[40]
Field testingCornDeep learningLaser emitterVision and odometry fusion[41]
Lab and field testingN/ADeep learningN/ATeleoperation and planning using sensor fusion[42]
Field testingN/ADeep learningRotating blade on a delta armRow guidance[43]
System overviewN/ADeep learning3 DOF manipulator with blade as end effectorN/A[44]
Lab testingSugarcaneImage processingRotating bladeVision-based guidance[45]
Table 4. Summary of research papers where the weeding tools used are the main focus.
Table 4. Summary of research papers where the weeding tools used are the main focus.
Type of StudyTarget CropWeed RecognitionWeeding ToolNavigationRef.
Lab testingPaddy fieldN/AA robot arm with a brush applies forcePotential method (attraction of repulsion of an obstacle)[46]
Field testingCarrotsN/AThe manipulator positions the weeding tool (tube stamp) via visual servoingRow following[47]
Field testingCottonImage processingArrow hoe, tine, cutting toolRow following[48]
Field testingPaddy fieldN/ASteering chains to stir the soilTeleoperation[49]
Field testingPaddy fieldCapacitive sensorWheels stir soil while movingCoverage navigation algorithm based on detected rice seedlings[50]
Lab and field testingMaizeDeep learningTwo vertically rotating discs with weeding knives.Conveyor belt[51]
Lab testingN/AN/A3 DOF robotic arm with glove-controlled cutterN/A[52]
Table 5. Summary of research where the environmental impact is investigated.
Table 5. Summary of research where the environmental impact is investigated.
Type of StudyTarget CropWeed RecognitionWeeding ToolNavigationRef.
Field testingPearsN/AThree razor-sharp pivoting bladesMap-based guidance[53]
Field testingSugar beetsImage processing/Deep learningBand-spraying and inter-row hoeingGPS-based positioning[54]
Field testingWheat and maiseN/ASprayer and flamingRow guidance with vision and RTK[55]
Field testingSugar BeetImage processingMechanical loosening/cutting and mulching/weed smothering with catch crops/thermal steamingGPS-based positioning[56]
Field testingEcological sugar beetImage processingCutting bladeGPS-based positioning[57]
System overviewN/AImage processingA laser-based weeding toolN/A[58]
Table 6. Overview of multi-robot approaches.
Table 6. Overview of multi-robot approaches.
DescriptionType of StudyTarget CropsWeed RecognitionWeeding ToolRef.
BoniRob with aerial droneSystem overviewSugar beetsDeep learningSprayer/weed stamping[59]
Three UGVs and two UAVsField testingMaize, wheat, olivesImage processingPatch spraying/air-blast sprayer/shallow soil tillage/thermal (burner)[61]
Multi-UGV system (AgBots)Simulation workN/AN/AN/A[62,63]
Human-robot cooperationField testingTomatoN/AIntra-row hoes[64]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lytridis, C.; Pachidis, T. Recent Advances in Agricultural Robots for Automated Weeding. AgriEngineering 2024, 6, 3279-3296. https://doi.org/10.3390/agriengineering6030187

AMA Style

Lytridis C, Pachidis T. Recent Advances in Agricultural Robots for Automated Weeding. AgriEngineering. 2024; 6(3):3279-3296. https://doi.org/10.3390/agriengineering6030187

Chicago/Turabian Style

Lytridis, Chris, and Theodore Pachidis. 2024. "Recent Advances in Agricultural Robots for Automated Weeding" AgriEngineering 6, no. 3: 3279-3296. https://doi.org/10.3390/agriengineering6030187

APA Style

Lytridis, C., & Pachidis, T. (2024). Recent Advances in Agricultural Robots for Automated Weeding. AgriEngineering, 6(3), 3279-3296. https://doi.org/10.3390/agriengineering6030187

Article Metrics

Back to TopTop