Search Results (29)

To address the challenges faced by heterogeneous Unmanned Aerial Vehicle (UAV) systems in complex task allocation, including over-reliance on centralized scheduling, training deadlock, inadequate capture of temporal collaboration, and unstable training under sparse reward conditions, this paper proposes a distributed task allocation algorithm based on reinforcement learning. The algorithm adopts a decentralized decision-making architecture, which enables the autonomous formation of UAV collaborative groups without the need for a global scheduling center. A cascaded submission timeout mechanism is introduced to prevent training deadlock; the combination of Long Short-Term Memory (LSTM) and attention mechanism is employed to accurately model temporal correlations and collaborative dependencies; and the Proximal Policy Optimization (PPO) algorithm is leveraged to optimize the training stability under sparse reward conditions. Experimental results demonstrate that the proposed algorithm achieves a 100% task success rate in scenarios of different scales, and its key metrics, including makespan, time cost and waiting time, are significantly superior to those of mainstream baseline methods such as the Genetic Algorithm (GA) and the Hungarian Algorithm (HA). Moreover, the algorithm still maintains excellent robustness under the conditions of UAV failures, parameter variations, and dynamic task perturbations. This method supports zero-shot generalization for any number of UAVs and tasks and provides an efficient and reliable solution for the real-time collaborative scheduling of heterogeneous UAV systems. Full article

Modern AMSs (automated manufacturing systems) on the one hand bring many benefits, but on the other hand, they are cumbersome to coordinate. AMSs consist of various subsystems (e.g., production lines) that share a finite number of resources (robots, machines, buffers, automated guided vehicles, etc.). This forces AMS designers to build flexible and decentralized systems. However, in these cases, the danger of deadlocks exists. Consequently, such a situation requires the application of advanced supervisors. One solution to the deadline problem is the application of Petri nets. This paper is motivated by AMS control based on deadlock prevention by means of ordinary Petri nets (OPNs) and generalized Petri nets (GPNs). This paper examines two areas of AMS Petri net-based model structures and presents methods of deadlock prevention. First, simpler structures of AMSs modeled by OPNs and GPNs will be investigated, and then more complex structures of AMSs modeled by the same kinds of Petri nets (PNs) will be analyzed. The siphon-based approach will be used for deadlock prevention in all of these cases. The principal results are introduced, explained, and illustrated through examples. Key results are introduced, especially in Example 1 and Example 2. Full article

An iterative method was developed in this study within a Petri net system (PNS) for flexible manufacturing systems (FMSs) to eliminate deadlocks. The algorithm employs variant tokens, ranging from a few to many, in idle places to identify subnet deadlocks through reachability states. Additionally, it utilizes the place invariant (PI) to resolve deadlocks. An algorithm with a simple example was developed and applied using a complex scenario. The developed algorithm is user-friendly and effectively eliminates deadlocks in the PNS of FMSs, producing optimal results. Full article

In the domain of application of PN theory, the system deadlock problem of a flexible manufacturing system (FMS) is a thorny problem that needs to be solved urgently. All the research has the same objective of designing optimal controllers with maximal permissiveness and liveness. Plenty of the past literature used deadlock prevention as the main control strategy that is implemented by control places. However, these methods usually forbid undesirable system states from being reached, while reducing the system’s liveness. This study employed the resource flow graph (RFG)-based method to achieve a deadlock recovery policy that can maintain maximal permissiveness by adding control transitions (CTs). Also, we improved the current definition of RFG and developed a systematic approach for generating the corresponding RFG, which is based on flow mirroring pair (FMP) functions and the software Graphviz 12.2.1. Furthermore, this study proposed an automatic method that forms DOT script for generating Graphviz images, which is convincingly demonstrated in this study to enhance the execution efficiency and recognition of circular waiting situations. Full article

With the rapid development of the economy, urban road congestion has become more serious. The travel times for vehicles are becoming more uncontrollable, making it challenging to reach destinations on time. In order to find an optimal route and arrive at the destination with the shortest travel time, this paper proposes a dynamic shortest travel time path planning algorithm with an overtaking function (DSTTPP-OF) based on a vehicular ad hoc network (VANET) environment. Considering the uncertainty of driving vehicles, the target vehicle (vehicle for special tasks) is influenced by surrounding vehicles, leading to possible deadlock or congestion situations that extend travel time. Therefore, overtaking planning should be conducted through V2V communication, enabling surrounding vehicles to coordinate with the target vehicle to avoid deadlock and congestion through lane changing and overtaking. In the proposed DSTTPP-OF, vehicles may queue up at intersections, so we take into account the impact of traffic signals. We classify road segments into congested and non-congested sections, calculating travel times for each section separately. Subsequently, in front of each intersection, the improved Dijkstra algorithm is employed to find the shortest travel time path to the destination, and the overtaking function is used to prevent the target vehicle from entering a deadlocked state. The real-time traffic data essential for dynamic path planning were collected through a VANET of symmetrically deployed roadside units (RSUs) along the roadway. Finally, simulations were conducted using the SUMO simulator. Under different traffic flows, the proposed DSTTPP-OF demonstrates good performance; the target vehicle can travel smoothly without significant interruptions and experiences the fewest stops, thanks to the proposed algorithm. Compared to the shortest distance path planning (SDPP) algorithm, the travel time is reduced by approximately 36.9%, and the waiting time is reduced by about 83.2%. Compared to the dynamic minimum time path planning (DMTPP) algorithm, the travel time is reduced by around 18.2%, and the waiting time is reduced by approximately 65.6%. Full article

Industry 5.0 aims to integrate humans and machines to achieve greater productivity, personalization, and sustainable development in the production process. Built on the foundation of Industry 4.0 which emphasizes automation, digitalization, and intelligent production processes, Industry 5.0 highlights the importance of human resources in modern manufacturing. Robotic arms have replaced traditional manpower, particularly in flexible manufacturing systems. However, integrating advanced machinery into workflows has increased competition in terms of securing resources, resulting in frequent deadlocks. Various deadlock prevention policies have been proposed to address this issue. Despite these efforts, resolving system deadlocks while achieving the optimal number of reachable states remains challenging. Based on existing research, we have developed a novel deadlock recovery method applicable to various flexible manufacturing systems. We designed an adaptable system and a controller that can restore the system to its fully operational state. Full article

The deadlock control problem in flexible manufacturing systems (FMSs) has received much attention in recent years. The formalism of the Petri net is employed to effectively model, analyze, and control deadlocks in an FMS case study. There are many kinds of deadlock prevention strategies based on the Petri net approach, where computational complexity is a major problem that needs to be considered. Based on the Petri net theory, this paper focuses on the two special subclasses in the ${\mathrm{S}}^3$PR net, namely the dual-process ${\mathrm{S}}^3$PR and the dual-process ${\mathrm{US}}^3$PR, in a bid to prevent deadlocks in an FMS. The relationship between the net structural characteristics and the deadlocks reached was analyzed, and then a regular method of adding controllers for these two models was proposed to reduce computational complexity. Full article

This paper is devoted to Petri net (PN)-based models of automated manufacturing systems (AMSs) with resources in order to prevent deadlocks in them. Their sustainability can be seen as the result of their deadlock freeness, leading to correct and fluent production, because AMSs with deadlocks work neither correctly nor fluently, need reconstruction and cause downtime in production. The paradigm of such PN models, S³PRs (systems of simple sequential processes with resources), is well known from the deadlock prevention point of view. Here, an extended S³PR (ES3PR) will be explored, with respect to its modelling and deadlock prevention. While in the case of S³PRs, ordinary Petri nets (OPNs) are used for these aims, here, for ES³PRs, generalized Petri nets (GPNs) are used. The reason for such a procedure is the possible presence of multiplex-directed arcs in the structure of PN models of AMSs. The significant alternation is that while, in the former case, the elementary siphons and dependent ones are sufficient for supervisor synthesis, here, in the later case, the GPNs and their siphons have to satisfy the max cs property. Full article

Automatic Guided Vehicles (AGVs) are widely used in flexible manufacturing systems for material handling inside the factory. Traffic management strategies, required to guarantee a conflict-free operation of the overall fleet, discretize the workspace of the AGVs and use the resulting graph model for route planning and execution. In zone control approaches, AGVs move from node to node on a permit basis, with limitations on the allowed number of AGVs at a time in each area of the graph to prevent and/or resolve deadlocks and conflicts. Hence, for an optimal implementation of traffic controllers in real manufacturing systems, it is essential to understand how the layout discretization influences the performance of the AGV network. This paper analyzes its effect in grid-like shaped workspaces by using a representative zone control algorithm and a recently developed improvement of it. Realistic numerical experiments on different layouts reveal that denser discretizations do not yield faster executions or increase in throughput, while lower control periods in the permit system entail significant performance uplifts. Full article

Controlling access to critical zones by drones is crucial for ensuring safety and efficient operations in various applications. In this research, we propose a strategy for controlling the access of a set of drones to a critical zone using timed automata and UPPAAL. UPPAAL is a model checker and simulator for real-time systems, which allows for the modeling, simulation, and verification of timed automata. Our system consists of six drones, a controller, and a buffer, all modeled as timed automata. We present a formal model capturing the behavior and interactions of these components, considering the constraints of allowing only one drone in the critical zone at a time. Timed automata are a powerful formalism for modeling and analyzing real-time systems, as they can capture the temporal aspects of system behavior. The advantages of using timed automata include the ability to model time-critical systems, analyze safety and liveness properties, and verify the correctness of the system. We design a strategy that involves signaling the approaching drones, preventing collisions, and ensuring orderly access to the critical zone. We utilize UPPAAL for simulating and verifying the system, including the evaluation of properties such as validation properties, safety properties, liveness properties, and absence of deadlocks. However, a limitation of timed automata is that they can become complex and difficult to model for large-scale systems, and the analysis can be computationally expensive as the number of components and behaviors increases. Through simulations and formal verification, we demonstrate the effectiveness and correctness of our proposed strategy. The results highlight the ability of timed automata and UPPAAL to provide reliable and rigorous analysis of drone access control systems. Our research contributes to the development of robust and safe strategies for managing drone operations in critical zones. Full article

Aiming to resolve the problems of slow convergence speed and inability to plan in real time when ant colony optimization (ACO) performs global path planning, we propose a path-planning method that improves adaptive ant colony optimization (IAACO) with the dynamic window approach (DWA). Firstly, the heuristic information function is modified, and the adaptive adjustment factor is added to speed up the algorithm’s convergence rate; secondly, elite ants and max–min ants systems are implemented to enhance the global pheromone updating process, and an adaptive pheromone volatilization factor is aimed at preventing the algorithm from enhancing its global search capabilities; then, the path optimization and withdrawal mechanism is utilized to enable smoother functioning and to avoid the deadlocks; finally, a new distance function is introduced in the evaluation function of DWA to the enhance real-time obstacle-avoidance ability. The simulation experiment results reveal that the path length of the IAACO can be shortened by 10.1% and 13.7% in contrast to the ACO. The iteration count can be decreased by 63.3% and 63.0%, respectively, leading to an enhanced optimization performance in global path planning and achieving dynamic real-time obstacle avoidance for local path planning. Full article

With the implementation of AGV technology and automated scheduling, storage and retrieval systems have become widely utilized in warehouse management. However, due to the use of unidirectional channels, AGV movement is restricted, and detours may occur frequently. Additionally, as the number of AGVs increases, deadlocks can arise, which lead to delays in order packaging and a decrease in overall warehouse performance. Hence, this paper proposes a dynamic scheduling method for task assignment and route optimization of AGVs to prevent collisions. The routing optimization method is based on an improved A* algorithm, which takes into account the dynamic map as input. Moreover, this paper investigates highly complex collision scenarios in bidirectional channels. Through simulation experiments, it is evident that scheduling methods based on bidirectional channels offer a clear advantage in terms of efficiency compared to those based on unidirectional channels. Full article

After the fourth industrial evolution, precision and automatic manufacturing have become increasingly widely accepted in production. With highly variable productivity and flexibility, flexible manufacturing systems (FMS) lower production costs and increase efficiency. Due to its resource shareability, unexpected system deadlock may occur in some specific situations. Many existing works use deadlock prevention as the primary control methodology in research on system deadlock control, while this type of control policy would constrain the transportation resources and reduce the system’s liveness. This paper adopts a new transition-based deadlock recovery policy as the direct control strategy, which uses generating and comparing aiding matrix (GCAM) to determine the optimal control transition. We also improve the existing GCAM-based method by reducing the computational redundancy. This kind of control strategy and its benefit could be demonstrated through two typical systems of simple sequential processes with resource (S³PR) nets and their Petri nets model. Full article

In the flexible manufacturing system deadlock prevention domain, researchers’ almost final target is to seek the maximally permissive controllers for solving the deadlock problems of flexible manufacturing systems. However, it seems a challenging work. Whatever you adopt, what kinds of methods, policies, and strategies, it seems complicated to obtain optimal controllers for deadlock prevention even if they claim their algorithms are optimal until the deadlock recovery is developed. Therefore, many experts, including us, have decided to design all kinds of actual maximally permissive recovery policies based on control transitions. It is a pity that these policies usually solve some particular flexible manufacturing systems’ deadlock problems. The controllers failed to recover the deadlock situation of flexible manufacturing systems once the idle or resource places were changed. In other words, these policies could never know or identify the real number of maximally permissive controllers in the same structure with different idle places. The precise combination of the number of idle places is called the saturated line, and it divides the reachability states into two region (saturated region and unsaturated region). Accordingly, this paper proposes a novel concept that achieves the precise maximum permissiveness and fits all kinds of flexible manufacturing systems. The experimental results show how our policy does in practice. Full article

Path planning is an important area of mobile robot research, and the ant colony optimization algorithm is essential for analyzing path planning. However, the current ant colony optimization algorithm applied to the path planning of mobile robots still has some limitations, including early blind search, slow convergence speed, and more turns. To overcome these problems, an improved ant colony optimization algorithm is proposed in this paper. In the improved algorithm, we introduce the idea of triangle inequality and a pseudo-random state transfer strategy to enhance the guidance of target points and improve the search efficiency and quality of the algorithm. In addition, we propose a pheromone update strategy based on the partition method with upper and lower limits on the pheromone concentration. This can not only improve the global search capability and convergence speed of the algorithm but also avoid the premature and stagnation phenomenon of the algorithm during the search. To prevent the ants from getting into a deadlock state, we introduce a backtracking mechanism to enable the ants to explore the solution space better. Finally, to verify the effectiveness of the proposed algorithm, the algorithm is compared with 11 existing methods for solving the robot path planning problem, including several ACO variants and two commonly used algorithms (A* algorithm and Dijkstra algorithm), and the experimental results show that the improved ACO algorithm can plan paths with faster convergence, shorter path lengths, and higher smoothness. Specifically, the algorithm produces the shortest path length with a standard deviation of zero while ensuring the most rapid convergence and the highest smoothness in the case of the shortest path in four different grid environments. These experimental results demonstrate the effectiveness of the proposed algorithm in path planning. Full article