Tetrahedral Mobile Robots: A Systematic Literature Review

Abstract

This paper presents a systematic review of the evolution and characteristics of mobile robots with a tetrahedral structure—a category of robots with high potential in fields such as space exploration, emergency response, and the inspection of hard-to-reach environments. The study is based on the examination of 27 relevant scientific articles selected from 4 international databases and aims to answer three essential research questions: what types of locomotion are used, what practical applications are targeted, and what challenges arise in the design of such systems. Following the quality assessment of the studies, a total of 20 articles were ultimately retained for the literature analysis. The analysis highlights a variety of locomotion strategies—among which rolling is the most common—as well as structural solutions tailored to specific use scenarios. Additionally, a range of technical challenges is identified, including stability control, the complexity of extension mechanisms, and sensor integration. This paper provides a clear overview of technological progress in the field and outlines useful directions for future research.

1. Introduction

Tetrahedral robots are mobile robots based on a symmetrical tetrahedral geometry, representing a three-dimensional solid composed of four triangular faces, which are either equilateral or formed by struts connected through nodes.

Although characterized by a simple structure, this category of robots is used in a wide range of fields, such as room inspection, space exploration in hazardous areas, rescue operations, and more.

The concept of a robot with a tetrahedral structure was developed to eliminate the limitations encountered during the overturning of conventional robots.

The main advantages of tetrahedral robots include stability provided by the robot’s symmetrical design, adaptability to various environments with complex obstacles due to their variable shape and size, and modularity.

Tetrahedral robots also present disadvantages such as low energy efficiency, slow rolling locomotion, and a complex control system, which limit their performance in dynamic applications.

The field of tetrahedral robots stands out from conventional robotics due to its potential to support various functional configurations. This approach provides a solid foundation for exploring novel mechanical and control solutions.

This category of robot, configured in the shape of a tetrahedron, although based on a rigid geometric structure, stands out through its high structural versatility due to

• Soft morphology [1];
• Modularity [2,3,4,5,6,7];
• Shape-shifting capability [8,9].

Thus, the tetrahedral shape does not limit functionality but rather provides a stable geometric framework for exploring complex robotic behaviors.

In addition to their structural diversity, tetrahedral robots also stand out through significant variation in terms of locomotion. They are equipped with various mobility mechanisms such as

• Crawling; through the actuation of movable elements, the robot can generate forces that move it in different directions [10];
• Rolling, by shifting its center of gravity; the robot can transition from one stable position to another [4];
• Resonance-based movement; by using oscillating mechanisms, the robot can achieve movement across different surfaces [11];
• Omnidirectional locomotion [9] achieved using unconventional locomotion units, such as simple universal wheels [12], active wheels [13], double wheels [14], mecanum wheels [15,16], spherical wheels [17], continuous alternative mechanisms [18,19], Laquos [17], and WESN [20].

This paper aims to identify the most relevant scientific contributions in the field of mobile tetrahedral robots and to analyze them from the perspective of proposed locomotion types, applicability, and encountered challenges. For this purpose, a systematic literature review (SLR) was conducted.

Through this approach, the paper aims to provide an in-depth understanding of the potential of tetrahedral robots in real-world applications.

2. Methodology

A systematic literature review (SLR) aims to provide a comprehensive overview of a topic studied.

It is a method used to identify, evaluate, and synthesize all relevant studies addressing a specific research question or set of research questions [21,22].

2.1. Research Questions

In the context of tetrahedral robots, the main research questions are presented in

Table 1. Research questions related to tetrahedral robots.

No.	Research Questions
RQ1	What type of locomotion is used?
RQ2	What practical applications are targeted by this category of robots?
RQ3	What are the most common challenges in their design and implementation?

.

2.2. Search Strategy

To conduct a broad and thorough search of the studies currently available in the specialized literature, we selected four databases: IEEE/IEL Electronic Library (IEL), Web of Science, Scopus Elsevier, and ScienceDirect. These are recognized as some of the most important sources in terms of domain diversity, their ability to provide up-to-date studies, and their quality in the field of engineering.

The extraction of relevant data consists of searching for scientific papers using specific terms such as “tetrahedral robot,” “tetrahedral mobile robot,” and “modular tetrahedral robot.” The selection targets papers published between 2005 and 2025, written in English.

• IEEE/IEL Electronic Library (IEL)—https://ieeexplore.ieee.org/
• Web of Science—https://www.webofscience.com/
• Scopus Elsevier—https://www.scopus.com/
• ScienceDirect—https://www.sciencedirect.com/

Following a preliminary analysis of the four databases using the previously mentioned keywords, 277 studies were identified in the IEEE/IEL Electronic Library (IEL). In the Scopus Elsevier database, 111 studies were found, 86 studies were identified in Web of Science, and 95 studies were found in ScienceDirect.

2.3. Inclusion/Exclusion Criteria

These criteria represent a set of predefined rules used to decide which of the identified studies will be analyzed and which will be excluded.

The criteria are intended to help the researcher focus on the studies that fall within the scope of the formulated research questions.

(a)

Inclusion criteria:

• IC1—The paper is written in English;
• IC2—The paper presents a tetrahedral, symmetrical structure;
• IC3—The paper is validated through a physical prototype;
• IC4—The paper is validated through publication in a scientific journal or conference;
• IC5—The locomotion system is non-conventional.

(b)

Exclusion criteria:

• EC1—The paper is not written in English;
• EC2—The robot’s structure does not exhibit tetrahedral geometry;
• EC3—It does not provide details regarding the physical implementation;
• EC4—It is not published in a validated (peer-reviewed) scientific source;
• EC5—The locomotion system is conventional.

The application of these criteria enabled a rigorous analysis of the identified studies, resulting in a total of 79 relevant scientific articles selected from the 4 databases.

A total of 18 studies from the IEEE/IEL Electronic Library (IEL), 19 from the Scopus Elsevier database, 12 from Web of Science, and 3 from the ScienceDirect database were excluded, as they did not meet the established criteria.

2.4. Quality Assessment of Studies Related to Tetrahedral Robots

Using the inclusion and exclusion criteria on the selected papers, a quality analysis will be conducted, applied only to those studies that meet the previously established criteria (

Table 2. Selection of papers following the quality assessment of the papers.

No.	Paper	C1	C2	C3	Final Rating
1.	[23]	Yes	No	Yes	Average
2.	[24]	Yes	Yes	Yes	High
3.	[25]	Yes	Yes	Yes	High
4.	[26]	Yes	Yes	Yes	High
5.	[27]	Yes	Yes	Yes	High
6.	[28]	Yes	No	Yes	Average
7.	[29]	Yes	No	No	Low
8.	[30]	Yes	Yes	Yes	High
9.	[31]	Yes	Yes	Yes	High
10.	[32]	Yes	Yes	Yes	High
11.	[33]	Yes	No	Yes	Average
12.	[34]	Yes	Yes	Partial	Average
13.	[35]	Yes	No	Yes	Average
14.	[36]	Yes	Yes	Yes	High
15.	[11]	Yes	Yes	Yes	High
16.	[37]	Yes	Yes	Yes	High
17.	[38]	Yes	Yes	Yes	High
18.	[39]	Yes	Yes	Yes	High
19.	[9]	Yes	Yes	Yes	High
20.	[40]	Yes	Yes	Yes	High
21.	[41]	Yes	Yes	Yes	High
22.	[42]	Yes	No	Yes	Average
23.	[43]	Yes	Yes	Yes	High
24.	[44]	Yes	Yes	Yes	High
25.	[45]	Yes	Yes	Yes	High
26.	[8]	Yes	Yes	Yes	High
27.	[46]	Yes	Yes	Yes	High

). This quality assessment aims to evaluate whether the studies provide reliable, relevant information supported by a physical model. Therefore, three selection criteria were considered:

(a)

Clarity of the purpose and objectives of the research (Criterion 1)

The paper provides coherent explanations regarding the purpose and objectives of the research, enhancing the relevance of the scientific approach.

(b)

Degree of applicability of the proposed solution (Criterion 2)

This criterion assesses whether the proposed solution can be implemented in a real-world context and has practical potential in the targeted environment.

(c)

Quality and relevance of the technical data presented (Criterion 3)

This criterion assesses whether the study includes concrete, measurable data that support the conclusions drawn.

Only the studies that meet all three criteria will be retained.

Following the quality assessment of the studies based on the three criteria, six studies were excluded as they did not meet one or more of the specified criteria.

The main criterion leading to the exclusion of papers is Criterion 2, which refers to the degree of applicability of the proposed solution.

The papers presented in [23,28,33,35,42], although they mention the existence of physical prototypes, do not provide practical data to validate their functionality in real-world environments, but only simulation results.

The paper in [29], although it mentions data related to tetrahedral robots within the modular architectures presented, addresses the subject of tetrahedral robots only superficially, without experimental validation. In contrast, the paper in [34] provides some data regarding experimental validation, but the technical information related to tetrahedral robots is limited, being discussed only in a subchapter that does not contain sufficient technical details, with the emphasis placed more on the control aspects of these modular robots.

2.5. Organization and Classification of Results

Following the search and selection process of the specialized literature, four databases (IEEE/IEL Electronic Library (IEL), Scopus Elsevier, ScienceDirect, and Web of Science) were investigated based on specific search criteria related to the chosen topic.

Although many papers were initially identified, not all of them were of interest, as they were not focused on tetrahedral robots.

Out of the 569 preliminary studies, 43 were excluded as duplicates.

Following further analysis, 447 of the remaining 526 studies were eliminated because they did not fall within the field of tetrahedral mobile robots.

Based on the inclusion and exclusion criteria, 52 of the 79 studies were removed, as they did not meet the five inclusion criteria outlined in Section 2.3

The final filtering stage involved the quality assessment of the remaining studies, which led to the exclusion of an additional seven articles.

The diagram in Figure 1 illustrates the paper exclusion process, from the preliminary phase to the final stage.

To provide a clear direction for the paper, three research questions were formulated, around which the entire study is structured.

These questions are presented in

Table 3. Research questions results.

No.	Research Questions	Selected Papers
RQ1	What type of locomotion is used?	[8,9,11,24,25,26,27,30,31,32,36,37,38,39,40,41,43,44,45,46]
RQ2	What practical applications are targeted by this category of robots?	[8,9,11,24,25,26,27,30,31,32,36,37,38,39,40,41,43,44,45,46]
RQ3	What are the most common challenges in their design and implementation?	[8,9,11,24,25,26,27,30,31,32,36,37,38,39,40,41,43,44,45,46]

, along with the selected papers that address them.

The final papers were organized based on the type of locomotion, year of publication, and country of origin.

It was observed that the most common type of locomotion is rolling locomotion, followed by omnidirectional locomotion (Figure 2). The year with the highest number of publications in the field is 2023, which suggests a growing and current interest in the researched topic (Figure 3). Regarding the country of origin, the United States, followed by China, holds the leading positions in terms of robotic innovation at the international level (Figure 4).

2.6. Literature Analysis

Following the review of the specialized literature [8,9,11,24,25,26,27,30,31,32,36,37,38,39,40,41,43,44,45,46], it was found that although tetrahedral mobile robots share a similar structure, they can be differentiated by their various types of locomotion, being adapted according to their intended purpose and operating terrain.

Therefore, we chose to classify these robots based on the type of locomotion used, identifying rolling locomotion, crawling, resonance-based locomotion, omnidirectional movement, or multi-modal locomotion—in cases where the robot combines two or more of the movement types (Figure 5, Figure 6, Figure 7 and Figure 8).

The analysis conducted for each paper includes details regarding the robot’s structure, the type of locomotion used, and information related to the experimental tests.

(a)

Rolling locomotion

One of the research directions regarding tetrahedral robots focuses on exploring modularity and deformability. Paper [24] addresses this topic by developing a modular robotic system capable of moving through rolling.

The proposed system is based on a tetrahedral configuration composed of extensible modules interconnected by joints (Figure 9). Locomotion is achieved using linear actuators present in each module, which can perform extension–contraction movements.

The mechanical design focuses on balancing flexibility and rigidity so that the robot maintains its shape during movement, even when its geometry is intentionally altered.

Experimental validation demonstrates the robot’s ability to perform repeated rolling cycles by shifting its center of gravity, both in a tetrahedral configuration and in other configurations.

The tests show the robot’s capability to navigate through obstacles or narrow spaces. The robot provides real-time visual feedback to the operator, facilitating control in various environments.

The robot is validated both through simulations, which confirm that the extension of the rods changes the center of gravity and generates controlled rolling, and through experimental tests that demonstrate the efficiency of the control system and the mechanical design.

The experimental tests highlighted an expansion ratio of 3:1, the change in angles from 60° to 26°, and the robot’s ability to climb a tube with angles ranging between 45° and 90°.

The authors note [24] that the main applications targeted by the robot include space exploration as well as search and rescue missions, due to the robot’s ability to adapt to various terrains.

Among the main design challenges identified are weight and scalability issues, the reliability of pressure sensors, and limitations related to the design itself.

Paper [27] proposes a tetrahedral-structured robot intended for everyday, military, and space exploration applications, capable of rolling locomotion through the extension of its structural struts (Figure 10).

In addition to the 6 extensible struts, the robot includes 4 flat nodes, a transmission mechanism for the extension/contraction of the rods, and a guiding and motion-limiting mechanism to prevent unwanted rod rotation.

The mathematical model of the robot analyzes its critical condition in three phases (extension, rolling, and impact) using a spring-damper model, where controlled extension over 4 s triggered rolling, in accordance with theoretical predictions.

Paper [27] also highlights the flexibility of movement direction control, where the authors demonstrate that after each rolling cycle, the robot can be programmed to select the next rolling axis.

The main design limitations include the need for a greater extension ratio, limited space for mounting and transmission, the requirement for a direction-changing mechanism, and the need for joints with two degrees of freedom to enable more complex movement.

The experimental tests focused on the robot’s behavior during a rolling cycle, during which the rod extension occurred over a duration of four seconds, after which the robot entered a free-fall phase.

Following the impact with the ground, the robot successfully returned to a stable configuration, demonstrating its potential for repetitive and controlled locomotion.

Among the tetrahedral robots with rolling locomotion is the deformable robot [30], composed of 4 rigid platforms and 6 chains actuated by 6 servomotors (Figure 11).

This robot is designed for applications involving rough terrains and exploration of irregular indoor or outdoor environments, such as grass, gravel, sand, or concrete.

To overcome obstacles, the authors propose three actions: simple rolling over obstacles (bridging), deformation to envelop or bypass the obstacle (enveloping), and active lifting over high obstacles where the first two methods cannot be applied (climbing).

Tests show that the robot can overcome obstacles up to 260 mm in height, as well as move across various terrains (gravel, sand, and concrete) at a speed of 0.2 m/s, confirming its efficiency.

It is noted that the design challenges of the robot stem from the need to combine rigidity with flexibility, as well as from increased collisions between the robot and the ground during movement errors.

Paper [31] presents an atypical structure compared to the previous ones, as it is composed of eight tetrahedra connected by eighteen rods and six nodes. The robot’s locomotion is based on the synchronized extension and contraction of multiple rods, which causes the robot to shift its center of gravity, thereby generating a rolling movement.

This robot is intended for applications such as search and rescue missions, autonomous space exploration, or monitoring of polar glaciers.

The experimental tests focused on validating motion control through the implementation of a PID controller, confirming the results obtained from simulations, where the feasibility of locomotion using the same algorithm was demonstrated. However, in practice, the experiments revealed lower travel speed and energy efficiency than those estimated in the simulations, due to the limited synchronization of the actuators.

The main challenges faced by the robot include the trade-off between flexibility and rigidity, as well as the limitation in the extension of the mechanisms, requiring two linear actuators per rod, which increases the robot’s complexity.

Paper [32] presents an approach like that of the previous study [31], both in terms of locomotion through strut extension–contraction to generate rolling motion and in the control principles applied. The difference between the two prototypes lies in the robot’s structure, which consists of 4 tetrahedra compared to the previous prototype, which is composed of 8 tetrahedra.

A diagram of the robot from paper [32] is shown in Figure 12.

The robot’s design faced challenges related to constructing joints capable of withstanding high forces, as well as the need for a precise and adaptable control system to enable proper movement under various terrain conditions. The robot was designed with the goal of adapting to any type of terrain.

Another relevant example is the robot presented in [40], which consists of four soft limbs made from pneumatic actuators mounted on a flexible frame, intended for applications such as search and rescue missions, agriculture, and the exploration of uneven terrain.

The type of locomotion used is rolling, achieved by shifting the center of gravity through sequential control of the curvature of the pneumatic limbs.

The “tumbling” strategy refers to the intentional shifting of the center of gravity using a combination of movements of the rear limbs, followed by a controlled correction of the robot’s orientation after each roll.

The experimental results show that the robot can follow linear and curved trajectories at pressures of 3 bar and frequencies between 0.65 and 0.90 Hz, achieving an average speed of 12.5 cm/s, compared to the simulations, which demonstrated that rolling becomes stable at frequencies greater than or equal to 55 Hz. The energy consumption analysis demonstrates that locomotion through rolling is 65% more efficient than crawling (as shown in a previous study [1]).

Unlike crawling locomotion, rolling involves both active and passive periods for each limb, meaning that not all limbs consume energy simultaneously, which leads to an overall reduction in energy consumption.

The challenges encountered relate to limitations of the pneumatic system and precise control of the limbs.

In a study on tetrahedral robots, the authors described in [41] a mobile robot with a deformable mechanism, with potential in applications requiring omnidirectional locomotion and adaptability to various terrains. It consists of four platforms connected by six URU (Universal-Revolute-Universal) kinematic chains, each equipped with an active rotary actuator.

The type of locomotion is based on rolling through symmetrical deformation. Two types of movement were planned: impact rolling (involving collisions with the ground) and non-impact rolling (where the trajectory of the center of mass is optimized to avoid impacts and reduce positioning errors).

The paper describes the testing of two prototypes of different sizes, which demonstrated precise movement on flat terrain as well as adaptability to various environments (obstacles, slopes, or uneven surfaces).

The first prototype, with a mass of 4 kg and dimensions of 550 × 620 × 540 mm, reached a speed of 41.25 mm/s, while the second prototype, with a mass of 10.22 kg and dimensions of 955 × 955 × 950 mm, was able to climb obstacles up to 180 mm high and ascend a ramp with an inclination of 25°.

The paper highlights challenges related to reducing movement errors, optimizing the trajectory, and synchronizing the URU mechanisms.

The paper from [8] presents a prototype focused on precise, impact-free movement, emphasizing accurate control and the avoidance of mechanical shocks.

The proposed robot is based on an extensible tetrahedral structure, built using Sarrus mechanisms, consisting of a central tetrahedron and four satellite tetrahedra (Figure 13).

The Sarrus mechanism allows the robot to convert rotational motion into linear motion, which is a major advantage in terms of geometric transformation.

The geometric transformation is presented in two forms: the first involves folding the robot like an umbrella, and the second refers to radial contraction (symmetrical folding toward the robot’s center), which significantly reduces the robot’s dimensions.

This robot is intended for traversing challenging terrains, even though the design process faced several challenges, including limitations related to adaptability, difficulty in controlling stability, and the complexity of kinematic modeling to ensure precise movement during rolling.

Experimental test results show that the robot is capable of performing controlled rolling motion, which supports its applicability in difficult environments, even though the testing was limited in terms of complexity.

The prototype confirmed the transition between the two configuration phases (from 0.46 m to 0.75 m in height) and managed to perform complete 90° rolls per cycle.

Another robot capable of changing its dimensions is the one presented in [46]. Its structure includes four nodes and six extensible RRR chains (two scissor-like elements capable of changing their length).

By rotating the motors, the scissor-like elements extend or contract radially to shift the center of mass, creating a controlled imbalance that causes the robot to topple onto a new face.

The structure operates in two phases (minimum and maximum), and tests confirm the robot’s ability to perform complete rolling in both configurations.

In the minimum configuration, the robot reaches a speed of 213 mm/s with a step length of 236 mm, while in the maximum configuration, the robot reaches a speed of 192 mm/s with a step length of 346 mm.

This variable-size robot is designed to adapt to complex environments, ranging from navigating through narrow spaces to traversing steep areas.

Maintaining balance and precision during deformation or rolling was the main challenge encountered during its design.

(b)

Crawling locomotion

An example of a crawling robot is presented in [25], designed for various environments such as sand, gravel, or inclined terrain. The applications target scenarios where adaptive mobility is essential, and the modular structure allows for rapid reconfiguration.

It consists of four tetrahedral cells made from adaptable rods actuated by electric motors and connected through linking nodes.

Locomotion is achieved through extension–contraction movements, controlled via commands sent from Matlab, using a PID control system implemented on a microcontroller, as demonstrated in the experimental tests.

The design of this robot involved challenges related to achieving an extension ratio of approximately 3:1, as well as issues concerning the nodes, which—besides needing to be strong—must also allow mobility during deformation and ensure stability.

Paper [39] presents a complex tetrahedral structure, with the robot intended for space exploration in extremely harsh environments. It is composed of a network of reconfigurable trusses made up of 12 tetrahedral cells and 26 active rods, equipped with sensors and microcontrollers, connected within an ART (Addressable Reconfigurable Technology) architecture.

Locomotion is achieved through “flopping” movement (rolling by imbalance) as well as complex reconfigurations, with the robot being capable of adapting to uneven terrain, carrying payloads, and altering its structural configuration to overcome vertical or inclined obstacles.

The tests mention the construction of three generations of robots (1-TET, 2-TET, and 12-TET) [39]. The 12-TET robot demonstrated a complete configuration, validated through MATLAB simulations and supported by position sensors, accelerometers, and wireless communication.

It has a mass of approximately 40 kg, telescopic struts with an extension ratio of up to 5.29:1, a full deployment time of 6–10 s, the ability to carry a 1 kg payload in the central node, and to climb slopes with inclinations of up to 40°.

In the development of these systems, the authors note challenges related to the robot’s high weight, unpredictable behavior during rod extension, and the fragility of components.

(c)

Resonance-based locomotion

Another noteworthy study involves a tetrahedral robot whose locomotion is based on the principle of resonance [11].

It consists of a rigid frame and a central sphere attached to the frame using four springs (Figure 14). The electronic components are mounted inside the sphere, which protects them from external factors.

The robot’s locomotion is achieved by exciting the natural frequencies of the structure through oscillations generated by internal motors that periodically pull on the springs, producing a swaying effect.

The robot is controlled by a controller that generates oscillations to create a sliding or jumping motion, causing the entire structure to lift off the ground. This allows it to move across surfaces such as tiles or carpet without the need for wheels or legs.

The simulation results showed a displacement of 56 cm in 7.5 s through the activation of structural resonance, while the experimental tests on the 2.17 kg prototype confirmed locomotion through jumps of up to 5 cm and a speed of 2.3 m/min, with an average energy consumption of 3 W and an estimated autonomy of 22 h.

The design faced challenges related to integrating resonance-based control into a closed structure, calibrating frequency and friction parameters to achieve controlled locomotion, and ensuring precise interaction between the robot and the ground in varied terrain conditions where wheels, legs, or other locomotion units typically fail.

(d)

Omnidirectional locomotion

The robot described in paper [38] consists of a central body and four locomotion modules equipped with spherical “Omni-Ball” units, like those in papers [9,43], but distinguished by a transformation mechanism that allows the robot to switch from a tetrahedral structure to a flat configuration.

The schematic design of the robots from papers [9,38,43] is presented in Figure 15.

The locomotion of all three robots is achieved using “Omni-Ball” units composed of a shaft that produces active movement via motors, and two passive rollers that enable omnidirectional displacement.

These locomotion units provide stability on inclined and uneven terrains [38], adaptability through structural transformation [9], and enhanced performance in obstacle traversal and shock absorption in harsh environments by using elastic materials and damping mechanisms [43].

Experimental tests confirm the core performance of the locomotion units through their ability for omnidirectional movement, climbing steps up to 24 mm high or an inclined ramp of 30° [38], the functionality of the transformation mechanism for adapting to narrow spaces of only 139.7 mm or climbing obstacles of 15.5 cm [9], and the robot’s efficiency in climbing higher steps and maintaining traction in critical positions [43].

All three papers [9,38,43] propose robots intended for use in search and rescue missions. The common design challenges among these robots include protecting the central body during impacts, maintaining stability on varied terrains, and addressing propulsion mechanism limitations by optimizing the locomotion units to avoid dead zones during movement.

Another representative example of robots equipped with omnidirectional locomotion units is presented in the study [44].

The paper describes a comparison between two tetrahedral robot prototypes: the first prototype, shown in Figure 16 (detailed in paper [45]), consisting of four triangular plates made of 2 mm thick plexiglass, assembled in the form of a triangular pyramid, and the second prototype, shown in Figure 17, composed of two 3D-printed structures called the skeleton and the core, with spherical locomotion units of the Omni-Ball type.

The locomotion of the first prototype is achieved using double universal wheel units, Figure 18a [44], and the robot is wirelessly controlled via a remote controller.

The locomotion of the second prototype is achieved using spherical Omni-Ball units, Figure 18b [44], and it is controlled through a wired controller.

The experimental tests confirm the functionality of both prototypes: the first was subjected to forward motion (runtime—10 s) and lateral motion (runtime—8 s) over a distance of 1 m, as well as testing on a ramp with a 20° inclination (runtime—22 s), while the second prototype was subjected to rolling movements performed manually, in order to demonstrate its ability to roll from any position on its three locomotion units at the base.

Both papers propose the development of tetrahedral robots intended for inspection or intervention in hard-to-reach environments, where omnidirectional maneuverability is essential.

It is highlighted that the design challenges are related to maintaining a stable center of gravity, optimizing the contact between the surface and the locomotion units, and selecting materials that provide sufficient rigidity without compromising the robot’s weight.

(e)

Multimodal locomotion

Another example of a system capable of rolling, crawling, and rotating locomotion is presented in papers [26,36], which are designed for inspection operations and search tasks in hard-to-reach environments.

These robots consist of a central pivot with four soft limbs mounted on it, actuated pneumatically (Figure 19).

The simulations in [26] showed that the tetrahedral robot with soft-type legs has a maximum speed of 2.48 mm/s, while the experimental tests confirmed the execution of crawling and rotational movements, but at lower speeds, below 2 mm/s.

The robot proposed in [26] aims to overcome the limitations of conventional soft robots, such as excessive volume, a high number of actuators, or low speed.

The robot in [36] provides real-time information through a console that maps joystick commands into circular trajectories of the limbs.

During development, both robots encountered similar challenges, such as maintaining stability, as well as distinct ones—such as the kinematic modeling of the soft limb in [26], and the integration of a video system capable of providing the operator with feedback for precise navigation in [36].

Another noteworthy study focuses on the Tetraflex robot [37], which consists of six pneumatic actuators arranged in a tetrahedral configuration, connected by four rigid nodes (Figure 20).

This design allows for controlled shape deformation through inflation and deflation of the limbs, enabling the robot to perform various types of movements such as rolling, crawling, or bounding, allowing it to adapt to narrow spaces or transport objects, making it suitable for exploring irregular environments.

Experimental tests in [37] show that rolling provides the highest accuracy (with a linear deviation of 2.3%), bounding achieves the highest speed (up to 19.6 mm/s), while crawling proves to be the most efficient for navigating narrow spaces, where the robot, (with a mass of 350 g) can capture objects of up to 100 g (≈30% of its own weight) and safely transport objects within its structure.

The challenges encountered relate to controlling the robot’s shape through pneumatic pressure, which leads to linear and angular deviations during movement or shape transformation.

3. Discussions, Challenges, and Future Research Directions

• Discussions

The literature analysis highlights several important aspects regarding the research on tetrahedral robots.

Rolling locomotion remains the most frequently studied strategy, validated through numerous experimental prototypes, but there is a growing interest in multi-nodal robots, which combine several types of movement, thus addressing the need for adaptability in complex environments.

Furthermore, a strong focus can also be observed on omnidirectional locomotion units (universal wheels, Omni-ball), reflecting a current trend toward achieving superior and more precise maneuverability.

At the same time, research on soft or reconfigurable structures, based on pneumatic actuators or extensible kinematic chains, suggests a transition toward more flexible robots, capable of modifying their shape to overcome obstacles or enter confined spaces.

A noteworthy direction is represented by resonance-based locomotion, a relatively new approach in the context of tetrahedral robots, which opens important perspectives even though experimental validation is currently limited.

However, most prototypes are analyzed and validated in controlled environments, which highlights the need for testing robots under complex conditions to confirm their potential.

In addition to the reviewed scientific literature, the practical interest and application potential of tetrahedral robots are also highlighted by the existence of patents aimed at the development of such systems [47,48,49,50,51,52].

b.

Challenges

Although the literature analysis points out that robots have high potential, they still face technological challenges regarding:

• Low energy efficiency and, consequently, limited autonomy;
• Flexibility;
• Structural complexity and symmetry make it difficult to integrate perception equipment since the robot can roll on any side, thus hindering the acquisition of a complete view of the surrounding environment;
• Maintaining stability during deformation.

c.

Future Research Directions

The analysis of the specialized literature highlights several research directions in the field of mobile tetrahedral robots, such as

• The use of smart materials such as shape-memory structures, soft actuators, or elastic materials;
• The application of artificial intelligence for trajectory planning or autonomous control, enabling the robot to adapt in real time to different situations;
• Integration of self-learning algorithms to enable robots to control their energy consumption;
• Self-reconfiguration in the case of modular or transformable robots, depending on the operating environment;
• Expansion of experimental testing in uncontrolled environments;
• Development of multimodal mechanisms capable of performing multiple types of locomotion or carrying out several tasks;
• Integration of video cameras and sensors to overcome the limitations imposed by the tetrahedral structure, through the addition of panoramic systems or sensors distributed on all faces.

4. Conclusions

The main objective of this paper is the systematic analysis of mobile robots with tetrahedral structures, based on information retrieved from four scientific databases: IEEE/IEL Electronic Library (IEL), Scopus, Web of Science, and ScienceDirect. The research was structured around three essential research questions: “What type of locomotion is used?”, “What practical applications are targeted by this category of robots?”, and “What are the most common challenges in their design and implementation?”

To answer the three research questions, the literature was selected through a multi-stage process. As a result, 27 scientific articles relevant to the chosen topic were identified, and following the quality assessment, 20 articles were retained as highly relevant.

The analysis results showed that rolling is the most frequently encountered form of locomotion in tetrahedral-structured robots. It was also observed that the year 2023 marked a peak in relevant publications in this field, with the United States leading the research effort based on the number of selected articles published.

The reviewed literature was structured according to the type of locomotion (rolling, crawling, resonance-based, omnidirectional, and multimodal movement), and the main topics addressed include operating principles, robot architecture, experimental testing, targeted applications, and design challenges.

The experimental tests presented in the analyzed papers validate the proposed models, highlighting how the robots perform movements across various types of surfaces. From the perspective of their intended applications, the robots are found in a wide range of contexts, from space exploration and search and rescue missions to agriculture, inspection, and more.

Regarding the limitations or challenges in design, it was found that most robots face issues related to balancing flexibility and rigidity, kinematic complexity, and stabilizing the center of gravity.

This study provides an overview of the progress in the field, synthesizing locomotion types, applications, and major challenges, and it may serve as a solid starting point for future research focused on functional optimization and the integration of artificial intelligence into these systems.