Robot cluster triangular lattice formation control method based on extremely simple local sensing

Abstract

The invention provides a cluster robot triangular lattice formation control method based on extremely simple local sensing and a cluster aggregation algorithm. By simplifying a sensing and decision-making mechanism, the method can significantly improve the self-organization capability, adaptability and low-cost deployment capability of the cluster robot, and is suitable for large-scale deployment scenes such as environment monitoring, post-disaster rescue, automatic patrol and the like. Specifically, the robot acquires position and state information of other robots in a neighborhood by using a local sensing sensor, forms a uniform equilateral triangle lattice formation through self-organization position adjustment, and keeps formation stability by using a simple interaction rule. After the tasks are completed, the robot achieves self-organizing aggregation through a cluster aggregation algorithm, and tight recovery or subsequent task allocation can be completed without global information or complex communication. According to the method, the hardware complexity and the communication requirement of the robot can be reduced, the method is flexibly adapted to a dynamic complex environment, and the method has good performance and wide applicability.

Description

Technical Field

[0001] The present invention belongs to the technical field of swarm robot control. Specifically, it relates to a self-organizing triangular lattice formation algorithm and a distributed photosensitive swarm robot deployment. Background Art

[0002] Swarm robot technology has gradually become a popular research field in the past two decades. With the progress of sensors, communication technologies, and computing capabilities, researchers have started to explore how to use a large number of simple robots to cooperate to complete complex tasks. Different from traditional single-robot systems that rely on complex control algorithms and powerful computing capabilities, swarm robots achieve efficient task execution through local perception and simple interaction rules. The self-organizing behavior among robots enables the swarm to adapt to dynamic environments without central control or global information, with high robustness, scalability, and flexibility, which makes swarm robots show great application potential in multiple fields such as warehousing, disaster rescue, and environmental monitoring. However, despite the broad application prospects of swarm robot technology, actual large-scale applications still face multiple challenges. First, traditional swarm robot systems usually rely on complex hardware, reliable communication networks, and efficient algorithms, which makes their deployment in real environments face high costs and technical difficulties. For example, in environments such as disaster rescue, robots must have high adaptability, and high-cost robots may not meet the requirements. Second, many existing algorithms require high sensor and computing capabilities of robots, which increases the complexity and cost of deployment and may lead to unstable performance. Especially when individual robots fail, it may affect the cooperation efficiency of the entire swarm.

[0003] To address these challenges, the present invention proposes a triangular lattice formation control method based on minimalist local perception, aiming to simplify system deployment by reducing the requirements for individual robot capabilities. This method enables robots to achieve effective lattice formation through self-organization without global information by designing simple local perception mechanisms and interaction rules, thereby reducing hardware requirements and communication complexity. This design not only enables swarm robots to flexibly adapt to dynamic environments but also improves the robustness and economy of the system, especially suitable for large-scale deployment. In addition, the present invention also innovates in the recovery mechanism after swarm robots complete tasks, proposing a local self-organizing aggregation algorithm. Through this mechanism, robots can re-concentrate after tasks, optimizing task reallocation and communication efficiency. In summary, the core innovation of the present invention is to promote the wide application of swarm robots in real environments through a minimalist perception and decision-making model, reduce hardware requirements, improve deployment efficiency, and provide a new direction for future robot miniaturization and large-scale deployment. Summary of the Invention

[0004] The objective of the present invention is to propose a triangular lattice formation control method for swarm robots based on minimalist local perception, aiming to achieve self-organization, low-cost, and large-scale deployment of robot swarms in dynamic environments by reducing the dependence on complex perception systems. This method can effectively solve the dependence on global information and communication during the self-organization formation and aggregation of swarm robots, enhancing the adaptability, efficiency, and robustness of swarm task execution, and is particularly suitable for tasks such as environmental monitoring and post-disaster rescue in large-scale robot swarms.

[0005] To achieve the above objective, the present invention proposes two core technical solutions: a swarm triangular lattice formation algorithm and a swarm aggregation algorithm. These algorithm designs combine a minimalist local perception model and simple interaction rules, enabling efficient cooperation of robot swarms without global information or complex communication.

[0006] The triangular lattice formation algorithm of the present invention is based on local perception and forms an equilateral triangle lattice structure by adjusting the distance and orientation between robots. The goal of this algorithm is to enable robots to execute tasks without relying on global information, but rather to achieve efficient formation through simple local perception and self-organization mechanisms. To this end, a perception model is established. Robots are equipped with multiple binary sensors that can detect the presence of neighboring robots within a limited perception range. Each robot obtains neighborhood information by perceiving the robots around it and adjusts its position according to predetermined interaction rules. Specifically, the sensor determines the relative position of the robot and neighboring robots and decides whether to adjust the distance. The data of the sensor outputs discrete three-valued perception: When a robot detects a neighbor within the perception range, it outputs 1, indicating the presence of a neighbor; When a robot detects a neighbor within a predetermined distance but without direct contact, it outputs 0; When there are no neighbors, it outputs -1.

[0007] The motion decision and algorithm steps are as follows: Based on the perception result, the robot calculates the relative distance to its neighbor. If the neighboring robot is far away, the robot will perform circular motion to adjust its position; If the distance between the robots is too close, it will rotate in place by adjusting the angular velocity until the desired distance is reached; As the robot continuously performs counterclockwise circular motion, its angular velocity and linear velocity will be adjusted according to random factors in the algorithm to ensure that the robots maintain an appropriate spacing. This design can effectively control the distance between the robot and its neighbor and gradually achieve triangular lattice formation; After introducing randomness, the movement of the robots will exhibit certain fluctuations. However, over time, the robot swarm will tend towards a stable state. Through long-term operation, the robots can stably maintain an equidistant relationship with their neighbors, forming a triangular lattice formation structure.

[0008] The swarm aggregation algorithm includes aggregation objectives and problems. After the swarm task is completed, the robots need to be recycled or reorganized for subsequent task allocation or robot system maintenance. The present invention proposes an aggregation algorithm based on local perception and a simple decision-making mechanism. Without global information and complex communication, the robots can automatically aggregate into a tight swarm.

[0009] The perception and obstacle avoidance mechanism is as follows: The robot obtains binary information on whether a neighbor is within a safe distance through a front perception sensor. If it senses another robot in front and the distance is less than the safe distance, the robot will stop moving or avoid it to prevent collision. Otherwise, the robot will continue with the aggregation action at the set speed.

[0010] The algorithm steps are as follows: The robot makes a preliminary perception to determine whether there are other robots around it. If there are, it adjusts according to the distance; if not, it continues to move; The robot will rotate in place by adjusting the angular velocity, or move along a circular trajectory to the aggregation position if there is no collision risk; By adjusting the movement trajectory, the robot will gradually move closer to the center of the swarm and finally form a tight aggregate. This process requires no global information and only relies on local perception; During the aggregation process, the robot will avoid collisions with other robots by sensing the safe distance in front. The sensor information will determine whether to trigger an avoidance behavior, thus ensuring the safety during the swarm aggregation process.

[0011] The advantages of this invention include: No global information is required. The algorithm of the present invention does not require global information or powerful computing capabilities. The robot only relies on local perception data for motion decision-making, which greatly reduces the system complexity and hardware requirements. Efficient swarm behavior. Through the interaction of local perception and simple rules, the robots can exhibit complex swarm behavior and efficiently execute tasks. This method shows high robustness and flexibility during task execution and is suitable for dynamic and uncertain environments. Low cost and easy deployment. Since the present invention reduces the dependence on high-precision sensors and complex communication systems, the hardware cost of the robots is greatly reduced, and the system is easier to deploy on a large scale, especially suitable for scenarios such as post-disaster rescue and environmental monitoring that require rapid response. Description of the Drawings

[0012] Figure 1It is the flowchart of the steps of the present invention; Figure 2 It is the perception model of the robot of the present invention; Figure 3 It is the simulation diagram of the cluster configuration process of the present invention at different field of view angles; Figure 4 It is the simulation diagram of the cluster aggregation process of the present invention at different field of view angles. Specific embodiments

[0013] The following further describes the present invention in detail with reference to the accompanying drawings and specific embodiments. For the step numbers in the following embodiments, they are only set for the convenience of elaboration and explanation, and no limitation is placed on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0014] S1 System architecture and hardware configuration. The cluster robot system in the present invention is based on a ground two-wheel differential drive robot platform and is applicable to various types of ground or wheeled robots. Specifically, each robot platform includes the following parts: Microcomputer: Each robot is equipped with a microcomputer, such as the NVIDIA Jetbot platform, running the Ubuntu operating system, which is used to run control algorithms, process perception data, and generate motion instructions.

[0015] Drive system: The robot moves through a two-wheel differential drive system. The two wheels respectively control the left and right turns of the robot, and the movement direction and speed of the robot are controlled by adjusting the rotational speeds of the two wheels.

[0016] Sensor system: Each robot is equipped with multiple perception sensors, including distance sensors, ultrasonic sensors, or infrared sensors. These sensors provide information about the distance and direction between the robot and other neighboring robots, which is used to support the relative position adjustment between the robots.

[0017] Power system: The robot uses a battery as the power source, and the battery capacity and endurance time are designed according to the requirements of the task.

[0018] In order to achieve triangular lattice formation and cluster aggregation, the robot is equipped with a variety of local perception sensors, including: Distance sensor: It is used to measure the distance to other robots. Each robot is equipped with at least three simple binary distance sensors denoted as , which can sense other robots in the front and side directions to ensure that the distance between the robot closest to itself and itself with a high degree of discretization is within a certain set boundary, and is used to form a triangular lattice formation.

[0019] Direction sensor: Used to determine the orientation of the robot and its relative direction to other robots. A simple gyroscope or a digital tilt sensor can be used as the direction sensor to determine the rotation angle of the robot.

[0020] Obstacle avoidance sensor: Used to detect whether there are obstacles in front to ensure that the robots do not collide during the clustering process. Based on ultrasonic or infrared technology, this sensor can judge the safe distance ahead in real time.

[0021] The specific sensing model is as Figure 1 shown. First, the sensor of the robot at can sense whether there are other robots within its sensing range . is a fan-shaped area with a radius of and a central angle of . is an arbitrary point on the plane, is the orientation of the robot at , , and is the angle between . For the sensor , the robot can sense whether there are other robots within its adjacent area at . is a fan-shaped area with a radius of and the same central angle as the sensor , which is , where .

[0022] Since collision between robots is considered in the clustering problem of the present invention, the sensor is designed to achieve obstacle avoidance for the robots. Specifically, the robot can sense whether there are other robots within its safe distance ahead, that is, , , where is the safe distance of the robot. To prevent collision between robots, during the operation of the robots, if the distance between the robots is less than , it is not desired for them to get closer. Therefore, takes a smaller value.

[0023] S2 Triangular Lattice Formation Control Method for Swarm Robots

[0024] Before the task starts, all robots are randomly deployed in the target area, and their initial positions are random. The initial layout does not require any specific arrangement, and the goal is to self-organize into a stable triangular lattice formation through a clustering algorithm.

[0025] Specifically, the formation process is as follows: Each robot adjusts its movement by sensing the positions and states of its surrounding neighbors and using simple local interaction rules. The specific steps are as follows: Sensing neighbors: The robot uses sensors to continuously sense the relative positions of other robots in its neighborhood. During the formation process, the robot only senses the neighbors that are relatively close to itself. If the distance between robots is less than a preset threshold, they are considered neighbors. The sensing module can sense whether there are other robots in its corresponding sensing area. At time step , define the set of robots in the area as , and the set of robots in the area as . The designed binary discretization sensing model can judge whether these sets are empty sets. When the set is empty, it indicates that there are no robots in that area; otherwise, it means there are robots. When robot senses that there are other robots in the neighboring area, that is, the area within a distance less than in its field of view, ; when robot senses that there are other robots within the sensing range but no robots in the neighboring area, ; when no robots are sensed, . In addition to the above three-valued input, the robot has no other information input, cannot perform relative positioning of neighbors, obtain global coordinates, or communicate between machines, and its motion decision only depends on .

[0026] Calculating relative positions: Each robot calculates the relative positions with neighboring robots based on the sensing information, including distance and direction.

[0027] Motion decision: Based on the relative positions, the robot calculates whether it needs to adjust its own position. If the distance from a neighbor is too far, the robot will move towards the neighbor to adjust the distance; if the distance is too close, the robot will adjust the angle by rotating to restore to an appropriate position.

[0028] Motion model: The robot adopts a linear velocity and angular velocity control model. The linear velocity controls the robot to move forward in a certain direction, and the angular velocity controls the robot to rotate around its center. Through these adjustments, the robot gradually approaches the target triangular lattice structure.

[0029] To ensure that the robots can stably form a triangular lattice formation, the present invention designs a control mechanism based on circular motion. Specifically: Each robot performs a counterclockwise circular motion for a relatively long time according to the perceived neighbor information. This circular motion can effectively adjust the relative distance between the robots, thereby helping the cluster to stabilize in the triangular lattice structure.

[0030] Each robot will decide whether to accelerate or decelerate its motion according to the positions of the neighboring robots, and adjust the relative angle with the neighbors to ensure that a uniform triangular layout is finally formed within the target area.

[0031] To avoid local aggregation or uneven distribution of the robot cluster during motion, the present invention enhances the adaptive ability of the system by introducing random factors. The angular velocity during the robot motion is designed as a value with a random adjustment term, ensuring that each robot can flexibly adjust its motion mode according to the changes in the local environment when performing circular motion. Randomness not only increases the flexibility of the system but also helps the robots avoid falling into a locally stable state, ensuring that the cluster can quickly form a stable triangular lattice structure.

[0032] During the process of the robots performing circular motion, each robot adjusts its motion through local perception information with its neighbors, ultimately ensuring the overall stability of the cluster. Each robot will adjust its speed and direction in real time according to the relative position and moving state of the neighboring robots, gradually forming an equilateral triangle formation structure. As Figure 2 shows the cluster configuration process under different field of view angles, and the cluster finally stabilizes into the configuration of a triangular lattice formation.

[0033] S3 Cluster Aggregation Algorithm

[0034] After the swarm robots complete their tasks, they usually need to perform swarm aggregation for task allocation, subsequent task execution, or robot recovery. The aggregation process needs to ensure that the robots can self-organize according to simple perception rules and motion decisions without global information.

[0035] The swarm aggregation process is divided into several stages, and each stage realizes the self-organizing aggregation of the robots through local perception and motion control: Perception process: Each robot judges whether there are other robots in front through sensors and measures the relative distance to them. According to the perceived results, the robot judges whether it needs to adjust its position. This part uses sensors and in combination to enhance the robot's environmental perception ability. Sensor can detect whether there are other robots within the front field of view, thereby promoting the mutual approach of the robots in the swarm. And sensor It is responsible for monitoring whether there are other robots within the safe distance ahead to design an effective obstacle avoidance mechanism. The two sensing modules used can sense whether there are other robots within the sensing areas corresponding to the modules. At time step , the set of robots defined within the area is , and the set of robots within the area is . The designed binary sensing model can determine whether the defined set is an empty set. When the defined set is an empty set, there are no robots in the area; otherwise, there are. When robot does not sense any other robots in its field of view or there are robots within the safe distance, it can safely perform the aggregation action. At this time, set ; if robot senses that there are other robots in its field of view but there are no other robots within the safe distance, then it does not need or cannot perform the aggregation action. At this time, .

[0036] Motion decision-making: Robots adjust their motion states according to the positions of neighboring robots sensed and the safe distance. If the distance between a robot and other robots exceeds the preset safe distance, the robot continues to move towards the center; if the distance is too close, the robot stops moving or performs obstacle avoidance.

[0037] Obstacle avoidance mechanism: During the aggregation process, robots monitor whether there are obstacles or collision risks through the front sensing sensors. When a collision risk is detected, the robot stops moving forward or changes direction until it ensures safety and then continues to move.

[0038] Aggregation and adaptive adjustment. During the cluster aggregation process, robots gradually move closer to the cluster center through circular motion or linear adjustment. To avoid collisions or over-concentration, the present invention designs an adaptive mechanism that enables robots to flexibly adjust their motion speeds to ensure that no conflicts occur during the aggregation process.

[0039] Circular motion: Each robot performs a counterclockwise circular motion around the cluster center to make appropriate relative position adjustments with its neighbors. Through this mechanism, robots can gather into a tight group within a short time.

[0040] Distance sensing and obstacle avoidance: During the aggregation process, robots avoid collisions or forming uneven aggregates by sensing the presence of surrounding neighbors and obstacles. Figure 3 shows the cluster aggregation process at different field of view angles.

[0041] To improve the efficiency and stability of the aggregation process, the robot makes decisions using discretized perception results, ensuring that each robot can quickly and effectively cooperate with its neighbors and ultimately form a stable cluster. The robot achieves a fast and stable aggregation process by continuously adjusting its motion state and avoiding obstacles.

[0042] The swarm robot system of the present invention realizes self-organization through local perception and simple interaction rules. Each robot automatically adjusts its motion state by perceiving the surrounding environment and gradually forms the required formation or swarm structure. This mechanism can reduce the robot's dependence on global information and improve the flexibility and robustness of the system.

[0043] The swarm robot system of the present invention demonstrates high robustness through local perception and simple rules. Even in the case of sensor failures or communication interruptions, the robot can still maintain the stability of task execution and swarm coordination. In addition, the robot can flexibly adjust its motion mode according to task requirements and environmental changes, demonstrating excellent adaptability.

Claims

1. A method for controlling the triangular lattice formation of a swarm of robots based on minimalist local perception, the method comprising: Providing a plurality of robots, each robot equipped with a local perception sensor for perceiving the positions and states of the robots within its neighborhood; Based on the perception results, each robot adjusts its relative position with the robots within its neighborhood according to predetermined interaction rules to achieve a triangular lattice formation. Characterized in that the method further comprises: The robots self-organize their positions through local perception information (including the relative distances and orientations of the robots within the neighborhood), so that the swarm of robots forms a uniform equilateral triangle formation in the local environment.

2. The method according to claim 1, wherein the local perception sensor is a binary sensor, and the binary sensor can perceive whether there are other robots within its field of view and make a motion decision according to the perception results.

3. The method according to claim 1, wherein each robot performs circular motion by controlling the linear velocity and angular velocity to ensure that the relative distance with the robots within its neighborhood remains within a predetermined range.

4. The method according to claim 1, wherein the robots adopt a random angular velocity adjustment mechanism to enhance the distance control between the robots, so that the swarm of robots can be stable in the desired triangular lattice formation.

5. The method according to claim 1, wherein the robots adjust the combination of the angular velocity and linear velocity to achieve dynamic stability with their neighbors and ensure the maintenance of the geometric structure of the triangular lattice without global information.

6. A method for aggregating a swarm of robots, the method comprising: Providing a plurality of robots, each robot equipped with a local perception sensor for perceiving its surrounding environment, and the sensor can perceive whether there are other robots within its front field of view; Based on the perception results, the robots judge whether there is a collision risk and avoid it. If there is no collision risk, they move closer to the center of the swarm; The robots perform self-organizing aggregation through local perception and simple decision rules to form a tight group. Characterized in that the method further comprises: The perception sensor is a binary sensor, and it decides whether the robot continues to move or avoids obstacles by perceiving whether the safety distance in front meets a preset condition.

7. The method according to claim 6, wherein the robots perform self-organizing aggregation by performing circular motion, and the robots adjust the angular velocity and linear velocity according to the safety distance information during the aggregation process.

8. The method according to claim 6, wherein the robots make motion decisions through the discretized perception results during the aggregation process, and the discretized results include: When the robot perceives the presence of a neighbor and the safety distance is appropriate, the robot moves closer to the center of the swarm; When the robot perceives the presence of a neighbor and the safety distance is insufficient, the robot avoids obstacles or stops moving.

9. The method according to claim 6, wherein the local perception sensor can detect whether the robot is within the safety distance. If it is within the safety distance, it stops moving; otherwise, it continues to perform the aggregation task.

10. The combined application of the triangular lattice formation control method according to claim 1 and the aggregation method according to claim 6, which is used for automatic formation and recovery in robot swarm tasks.

Images