A cooperative robot gas source positioning method based on a bionic algorithm

Abstract

The application discloses a kind of based on bionic algorithm's cooperative robot gas source positioning method, belong to detection positioning field.The present application uses the rule that multiple small circles cover large circle, carries out the initial time placement of robot, realizes the maximum coverage to space;While reasonably allocating the detection space of single robot, eliminate space omission, so that robot cluster system can be contacted with target gas diffused in space with higher efficiency;Adopt the bionic gas source positioning method based on concentration gradient, drive robot constantly move to the direction of gas source leakage concentration high;And adopt the cooperative rule of robot cluster system suitable for bionic algorithm, the cooperative work of robot cluster is synchronized, both retain the advantages of low algorithm calculation cost, also make robot cluster system can be positioned gas source with higher efficiency and higher robustness quickly.The present application is suitable for gas detection field, for improving the efficiency and robustness of robot cooperative gas source positioning.

Description

Technical Field

[0001] This invention relates to a gas source localization method for collaborative robots based on a biomimetic algorithm, belonging to the field of detection and localization. Background Technology

[0002] In today's society, understanding the composition, concentration, and harmful properties of the air around us is considered a crucial task. Meanwhile, robots have become an integral part of our daily lives; in home and workplace environments, mobile robots with sensing capabilities can monitor ambient gas concentrations. This is particularly desirable in many different applications, including security, surveillance, mine clearance, and search and rescue. To endow these robots with sensing capabilities, gas sensing devices need to be integrated into these robotic platforms. These gas sensors should be able to detect a wide variety of gases, exhibit high sensitivity to these gases, and respond quickly and robustly. In this way, mobile robots can perform many gas-related tasks, among which active localization is a popular application.

[0003] Currently, mobile robot source localization methods are mainly divided into four categories: boundary tracking algorithms, probabilistic Bayesian inference methods, information path planning strategies, and biomimetic algorithms. Each of these four types of localization methods has its own significant shortcomings. For example, Bayesian source localization methods have wide applicability, but require prior information about the source parameters; information path planning strategies can obtain more accurate source location and intensity estimates, as well as faster search times, but due to the iterative and optimization process, the computational cost is high; while biomimetic algorithms have low computational cost, are easy to implement, and perform well in concentration fields with well-defined gradients, but they also suffer from lower accuracy and efficiency.

[0004] In nature, social animals can accomplish complex cooperative behaviors through simple collaboration among individuals. By studying and applying this phenomenon, autonomous vehicle swarms can achieve complex tasks through simple interactions within unmanned workshops, thereby improving their efficiency and robustness. Summary of the Invention

[0005] To address the problem of low efficiency in the localization of gas sources in existing technologies, the main objective of this invention is to provide a collaborative robot gas source localization method based on a biomimetic algorithm. By combining the initial position placement rules of multiple robots and robot cooperation rules with a gas source localization method based on concentration gradients, the efficiency and robustness of collaborative robot gas source localization are improved.

[0006] The objective of this invention is achieved through the following technical solution:

[0007] This invention discloses a biomimetic algorithm-based collaborative robot gas source localization method. At the initial moment of gas source localization, a robot swarm is deployed using a maximum coverage placement rule, enabling the robot swarm system to completely cover the area to be detected, eliminating any gaps and improving detection efficiency. A concentration gradient-based biomimetic gas source localization method drives the robots to continuously move towards areas with high gas source leakage concentration. Furthermore, a robot swarm system collaboration rule suitable for biomimetic algorithms is adopted to synchronize the collaborative operation of the robot swarm. This retains the advantages of the original algorithm, such as low computational cost, while also enabling the robot swarm system to quickly locate the gas source with higher efficiency and greater robustness.

[0008] This invention discloses a gas source localization method for cooperative robots based on a biomimetic algorithm, comprising the following steps:

[0009] Step 1: Arrange the robot cluster using the maximum coverage placement rule so that the robot cluster system can contact the target gas more efficiently.

[0010] The space where the gas source leaks can be considered a large circular space, and each individual robot has its own area of ​​operation, thus each individual robot can be considered a small circle. The rule of using multiple small circles to cover the large circle achieves maximum coverage of the space.

[0011] When the number of robots is less than 5, a method of dividing the circumference equally based on the pigeonhole principle is adopted. Each small circle must cover the circumference of the large circle if it is completely covered. Let the number of small circles be n, their radius be r, and the radius of the large circle be R. The method of dividing the circumference equally uses n small circles to divide the circumference of the large circle into equal parts. The chords corresponding to the arcs after division are used as the diameters of each small circle, achieving the minimum radius of the small circles while effectively covering the large circle.

[0012] When n=2, each small circle needs to cover Given the circumference of the large circle, it is easy to see that the small circle completely coincides with the large circle, that is, the radius of the small circle is equal to the radius of the large circle, r = R;

[0013] When n > 2, the radius of the small circle is:

[0014]

[0015] When the number of robots is greater than or equal to 5, using the method of dividing the circumference equally will result in smaller circles not covering the larger circle. In this case, the method of covering the circumference with an isosceles triangle is used. That is, the n smaller circles intersect at the center of the larger circle, the circumference of the larger circle is divided into n equal parts, and the lines connecting the two endpoints of the chord corresponding to each arc segment to the center of the larger circle form the legs of an isosceles triangle, the chord is the base of the isosceles triangle, and the smaller circle is the circumcircle of the isosceles triangle. The radius of the smaller circle is:

[0016]

[0017] For any number of robots, there exists a universally applicable, relatively optimal solution that completely covers the gas diffusion space: placing each robot at the center of a small circle achieves maximum coverage. With maximum spatial coverage, the robot swarm system can reach the target gas with greater efficiency.

[0018] Step 2: Employ a biomimetic gas source localization method based on concentration gradient and use multi-robot cooperative rules to drive the robot cluster to efficiently approach the gas source.

[0019] 2.1 A biomimetic gas source localization method based on concentration gradient is adopted to drive the robot to approach the gas source, which has low computational cost and high economic efficiency in planning trajectory.

[0020] The robot consists of an operating system for data processing and a drive system for operation. A gas concentration sensor is mounted on each side of the robot, and these sensors are cross-connected to servo motors on both sides. The output of these sensors directly drives the servo motors. The relationship between the angular velocity of the servo motors and the gas sensor output is as follows:

[0021]

[0022] Among them, M L and M R c1 and c2 represent the angular velocities of the robot's left and right servo motors, respectively; c1 and c2 are coefficients whose values ​​can be adjusted according to the environment; S R and S L These are the measurement outputs from the gas sensors on the right and left sides of the robot, respectively. When the concentration on the left is higher than on the right, the angular velocity of the right servo motor is higher than that on the left, causing the robot to rotate towards the side with the higher concentration, thus approaching and locating the gas source. During this process, the operating system simultaneously monitors whether the measured concentration value is increasing; if it is decreasing, the robot turns around. This addresses the problem of the robot straying far from the gas source when using a concentration gradient-based gas source localization algorithm.

[0023] 2.2 Adopting collaborative rules in robot clusters improves the overall efficiency of robot clusters.

[0024] The coordination rules between robot clusters include three communication states:

[0025] 1) No communication: At the initial moment, each robot moves according to the gas source localization method based on the concentration gradient at its initial distribution position;

[0026] 2) Attraction: When one of the robots detects gas and moves closer to the gas source using a gas source localization method based on concentration gradient, it will send an attraction signal to other robots that have not yet detected gas or have not yet moved closer to the gas source, causing other robots to move closer to that robot, thereby improving the efficiency of other robots in detecting gas and locating gas sources.

[0027] 3) Stop: When a robot locates a gas source, it will send a stop signal to other robots that are still probing, causing them to stop probing and enter a hibernation state, or it will allow other robots to come directly to the robot that sent the signal, that is, to bring all robots to the vicinity of the gas source to carry out the next operation.

[0028] Step 3: The gas sensor on the robot can measure the gas concentration at the robot's location. When the gas concentration exceeds the set threshold, it is determined that the robot has successfully located the gas source, thus realizing the location of the leaking gas source.

[0029] Beneficial effects:

[0030] 1. This invention discloses a collaborative robot gas source localization method based on a biomimetic algorithm. It employs a rule of multiple small circles covering a large circle for initial robot placement, achieving maximum spatial coverage. Simultaneously, it rationally allocates the detection space for each robot, eliminating spatial omissions, enabling the robot swarm system to contact the target gas diffused in space more efficiently, thereby improving the efficiency of robot gas source localization.

[0031] 2. The present invention discloses a biomimetic algorithm-based collaborative robot gas source localization method. In the process of driving the robot to approach the gas source, a biomimetic gas source localization algorithm based on concentration gradient is adopted. At the same time, the collaborative rules between the robots are added. That is, the original algorithm retains the advantage of low computational cost, and also enables the robot swarm system to quickly locate the gas source with higher efficiency and higher robustness. Attached Figure Description

[0032] Figure 1 This is a flowchart of a collaborative robot gas source localization method based on a biomimetic algorithm disclosed in this invention;

[0033] Figure 2 This is a schematic diagram illustrating the maximum coverage of the three-robot circumferential division method in a collaborative robot gas source localization method based on a biomimetic algorithm disclosed in this invention.

[0034] Figure 3 This is a schematic diagram illustrating the maximum coverage of the four-robot circumferential division method in a collaborative robot gas source localization method based on a biomimetic algorithm disclosed in this invention.

[0035] Figure 4This is a schematic diagram illustrating how the five-robot circumferential division method cannot cover the gas space in a collaborative robot gas source localization method based on a biomimetic algorithm disclosed in this invention.

[0036] Figure 5 This is a schematic diagram illustrating the maximum coverage of the five-robot isosceles triangle method in a collaborative robot gas source localization method based on a biomimetic algorithm disclosed in this invention.

[0037] Figure 6 This is a schematic diagram illustrating how each robot in this invention performs gas source localization based on a concentration gradient.

[0038] Figure 7 This is a schematic diagram illustrating how the robot that detected the gas sends an "attraction" signal to other robots. Detailed Implementation

[0039] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. The technical problems solved by the present invention and its beneficial effects are also described. It should be noted that the described embodiments are only intended to facilitate understanding of the present invention and do not constitute any limitation thereof.

[0040] This embodiment discloses a collaborative robot gas source localization method based on a biomimetic algorithm, specifically for locating and tracking CO2 gas leaks in the environment. The specific implementation steps are as follows:

[0041] Step 1: Deploy the robot swarm using the maximum coverage placement rule to enable the robot swarm system to detect contact CO2 gas more efficiently.

[0042] In gas source localization methods, the initial step involves using mobile robots equipped with gas sensors to detect gases in the contact space. Clearly, in a multi-robot swarm system, to achieve a rapid and effective gas source localization process, we cannot place all robots at the same starting point or randomly. Initially, using appropriate placement rules to scatter the robots in suitable locations can effectively improve the efficiency of the robot swarm system in detecting the target gas.

[0043] In this embodiment, initially, a CO2 gas leak causes CO2 gas to be distributed throughout the space. To achieve maximum coverage of the space, the unmanned vehicles are placed at different initial locations within the space, with the placement of the unmanned vehicles following the rules below:

[0044] This embodiment uses five unmanned vehicles (UAVs) for gas source localization, with the effective radius of each UAV set as r and the leakage space radius of the CO2 gas source as R. Since there are five UAVs, as... Figure 4As shown, the method of dividing the circumference equally can no longer guarantee maximum coverage of the space. Therefore, we adopt the isosceles triangle method with circular coverage, such as... Figure 5 As shown: Five smaller circles intersect at the center of a larger circle. The circumference of the larger circle is divided into five equal parts. The lines connecting the two endpoints of the chord corresponding to each arc segment to the center of the larger circle form the legs of an isosceles triangle, with the chord being the base of the isosceles triangle. The smaller circles are the circumcircles of the isosceles triangles. The radii of the smaller circles are:

[0045]

[0046] By placing each autonomous vehicle at the center of the small circle at the initial moment, the autonomous vehicle cluster system can achieve maximum coverage of the CO2 gas source leakage space, thereby enabling it to access the CO2 gas more efficiently.

[0047] Step 2: Employ a biomimetic gas source localization method based on concentration gradient and use multi-robot cooperative rules to drive the robot cluster to efficiently approach the gas source.

[0048] 2.1 A biomimetic gas source localization method based on concentration gradient is adopted to drive the robot to approach the gas source, which has low computational cost and high economic efficiency in planning trajectory.

[0049] The main components of each unmanned vehicle should consist of an operating system for data processing and a drive system for driving. The operating system mainly includes a data processing part, including a microcontroller, and the drive system mainly includes a drive part, including servo motors. The necessary sensors for the unmanned vehicle should include a CO2 gas concentration sensor mounted on each side. The output of the CO2 gas concentration sensors on both sides of the unmanned vehicle directly drives the servo motors on both sides, and is cross-connected with the servo motors on both sides. The relationship between the angular velocity of the servo motors and the relationship between the output of the gas sensors are shown in equation (3).

[0050] When the CO2 gas concentration on the left is higher than that on the right, the angular velocity of the right servo motor is higher than that on the left. The unmanned vehicle then turns to the left (towards the higher concentration) to approach and locate the gas source. During this process, the system simultaneously monitors whether the measured concentration is increasing; if it is decreasing, the unmanned vehicle turns around.

[0051] 2.2 Add collaborative rules to the robot swarm system to improve the overall efficiency of the robot swarm.

[0052] The coordination rules between the autonomous vehicles include the following three states of communication signals:

[0053] 1) No communication. Initially, each autonomous vehicle begins its gas source localization algorithm based on concentration gradients according to its initial distribution position. At this time, there is no communication between the vehicles. Figure 6 As shown;

[0054] 2) Attraction. When one of the autonomous vehicles detects CO2 gas and continuously approaches the CO2 gas source using a concentration gradient-based gas source localization algorithm, it will send an "attraction" signal to other autonomous vehicles that have not yet detected CO2 gas or approached the CO2 gas source. This causes other autonomous vehicles to move towards that vehicle, improving the efficiency of other autonomous vehicles in detecting and locating CO2 gas sources. Figure 7 As shown;

[0055] 3) Stop. Once a drone has located a CO2 gas source, it will send a "stop" signal to other drones that are still probing, causing them to stop probing and enter a dormant state, or it will bring other drones directly to the drone that sent the signal, meaning all drones will be around the CO2 gas source to proceed with the next step.

[0056] Step 3: Determine if the unmanned vehicle has reached the CO2 gas source

[0057] The CO2 gas concentration at the location of the unmanned vehicle is measured by a CO2 gas sensor installed on the vehicle. When the CO2 gas concentration exceeds the set threshold of 2000ppm, the unmanned vehicle is considered to have successfully located the CO2 gas source.

[0058] In summary, the present invention enables unmanned vehicles to access CO2 gas more quickly by maximizing the coverage of the gas space in step 1; and by combining a biomimetic gas source localization algorithm based on cooperative rules in step 2, compared with the original biomimetic algorithm based on concentration gradient, the multi-unmanned vehicle cluster system not only retains the advantages of low cost and low computational load of the biomimetic algorithm, but also significantly improves the efficiency and success rate of locating the CO2 gas source.

[0059] The above detailed description further illustrates the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is merely a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for locating gas sources in a collaborative robot based on a biomimetic algorithm, characterized in that: Includes the following steps, Step 1: Arrange the robot cluster using the maximum coverage placement rule so that the robot cluster system can contact the target gas with higher efficiency; The space where the gas source leaks is considered as a large circular space, and each robot has its own area of ​​operation, so each robot is considered as a small circle; the rule of using multiple small circles to cover the large circle can achieve maximum coverage of the space. When the number of robots is less than 5, the circumference is divided equally based on the pigeonhole principle. The premise that each small circle covers the large circle is that it must cover the circumference of the large circle. Let the number of small circles be n, the radius be r, and the radius of the large circle be R. The circumference division method is to use n small circles to divide the circumference of the large circle equally. The chord corresponding to the arc after division is used as the diameter of each small circle. This can obtain the smallest radius of the small circle while effectively covering the large circle. When n=2, each small circle needs to cover Given the circumference of the large circle, it is easy to see that the small circle completely coincides with the large circle, that is, the radius of the small circle is equal to the radius of the large circle, r = R; When n > 2, the radius of the small circle is: When the number of robots is greater than or equal to 5, using the method of dividing the circumference equally will not allow the smaller circles to cover the larger circle. In this case, the method of covering the isosceles triangle is used; that is, the n smaller circles intersect at the center of the larger circle, the circumference of the larger circle is divided into n equal parts, and the lines connecting the two endpoints of the chord corresponding to each arc segment to the center of the larger circle form the legs of the isosceles triangle, the chord is the base of the isosceles triangle, and the smaller circle is the circumcircle of the isosceles triangle; the radius of the smaller circle is: For any number of robots, there is a relatively optimal solution that is universal and can completely cover the gas diffusion space, which means that placing each robot at the center of the small circle can achieve maximum coverage of the space; when the space is maximized, the robot swarm system can reach the target gas with higher efficiency. Step 2: Employ a biomimetic gas source localization method based on concentration gradient and use multi-robot cooperative rules to drive the robot cluster to efficiently approach the gas source; 2.1 A biomimetic gas source localization method based on concentration gradient, which has low computational cost and high economic efficiency in trajectory planning, is adopted to drive the robot to approach the gas source; The robot consists of an operating system for data processing and a drive system for operation. A gas concentration sensor is mounted on each side of the robot, and these sensors are cross-connected to servo motors on both sides. The output of these sensors directly drives the servo motors. The relationship between the angular velocity of the servo motors and the output of the gas sensors is as follows: in, M L and M R c1 and c2 represent the angular velocities of the robot's left and right servo motors, respectively; c1 and c2 are coefficients whose values ​​can be adjusted according to the environment; S R and S L The measurement outputs are from the gas sensors on the right and left sides of the robot, respectively. When the concentration on the left side is higher than that on the right side, the angular velocity of the right servo motor is higher than that on the left side, so the robot turns towards the side with higher concentration to approach and locate the gas source. During this process, the operating system monitors whether the measured concentration value is increasing. If it is decreasing, the robot turns around to solve the problem of the robot moving away from the gas source under the gas source localization algorithm based on concentration gradient. 2.2 Adopting collaborative rules in robot swarms to improve the overall efficiency of robot swarms; The coordination rules between robot clusters include three communication states: 1) No communication: At the initial moment, each robot moves according to the gas source localization method based on the concentration gradient at its initial distribution position; 2) Attraction: When one of the robots detects gas and moves closer to the gas source using a gas source localization method based on concentration gradient, it will send an attraction signal to other robots that have not yet detected gas or have not yet moved closer to the gas source, causing other robots to move closer to that robot, thereby improving the efficiency of other robots in detecting gas and locating gas sources. 3) Stop: When a robot locates a gas source, it will send a stop signal to other robots that are still probing, causing them to stop probing and enter a hibernation state, or it will cause other robots to come directly to the robot that sent the signal, that is, to bring all robots to the vicinity of the gas source to carry out the next operation. Step 3: The gas sensor on the robot can measure the gas concentration at the robot's location. When the gas concentration exceeds the set threshold, it is determined that the robot has successfully located the gas source, thus realizing the location of the leaking gas source.

Images