CN114839990A - Cluster robot experiment platform - Google Patents

Abstract

The invention is a cluster robot experiment platform, which belongs to the field of robot technology. The experimental platform of the present invention includes a cluster robot, an experimental site, a communication system, a positioning system and a host computer. The swarm robot is a desktop mobile robot with two-wheel differential drive and moves in the experimental site. The communication system is equipped with a wireless communication module on the host computer and each robot to form a wireless ad hoc network. The positioning system includes a depth camera and a positioning program. The depth camera is installed just above the center of the experimental site, and the positioning program calculates the position of each swarm robot. The host computer sends control commands to a single robot or the entire group of robots, and displays the robot's running status. The invention integrates a robot, a communication system, a positioning system and a host computer, and has the conditions for centralized and distributed clusters to complete experiments such as formation control and area coverage, which facilitates the transplantation and verification of the cluster control algorithm and is beneficial to the improvement of the cluster control algorithm. and assessment.

Description

A swarm robot experiment platform

technical field

The invention belongs to the technical field of robots, and in particular relates to a cluster robot experiment platform.

Background technique

At present, swarm robots have become one of the hot research fields of robotics, and many swarm self-organizing control algorithms have emerged. Considering the cost, time-consuming and complexity of physical experiments, especially for large-scale swarms and underwater swarms, most researchers can only verify the effectiveness of their control algorithms through theoretical proofs and simulations. However, theoretical calculations and simulations are difficult to accurately simulate the interaction between real robots and robots, between robots and the environment, and the effects of environmental noise and interference, sensor errors, communication delays, and other factors on the performance of the control method. Therefore, in order to realize the experimental verification of the swarm control algorithm, it is necessary to propose a swarm robot experiment platform.

SUMMARY OF THE INVENTION

In order to make up for the deficiencies of the prior art, the present invention provides a swarm robot experiment platform, which has functions such as swarm positioning, communication, control and status information display, supports experiments of centralized and distributed swarm robots, and is helpful for verification The effectiveness of the cluster control algorithm.

The invention provides a swarm robot experiment platform, which includes a swarm robot, an experiment site, a communication system, a positioning system and a host computer. The swarm robot is a desktop mobile robot with two-wheel differential drive, which moves in the experimental site. The robot itself carries a distance sensor. The communication system is as follows: the host computer and each cluster robot are equipped with wireless communication modules to form a wireless ad hoc network. The positioning system includes a depth camera and a positioning program. The depth camera is installed just above the center of the experimental site and is connected to the upper computer. The positioning program calculates the pose information of each swarm robot and transmits it to the upper computer. The host computer issues the control commands of a single robot or the entire group of robots, and displays the robot running status.

In the positioning system, the positioning program uses the images captured by the depth camera to perform external auxiliary positioning of the robot.

In the positioning system, the positioning program is provided with two robot self-positioning methods, one is to use the angle and speed output by the gyroscope and encoder carried by the robot itself, and the current position of the robot is obtained by accumulative calculation; When the direction of the robot is parallel to the X axis or the Y axis of the global coordinate system, the current position is obtained by measuring the distance to the surrounding wall of the experimental site through the distance sensor carried by the robot itself.

The upper computer is composed of an individual control module and a cluster control module. In the individual control module, the host computer establishes communication with any robot in the cluster, and sends the target pose, tracking path and target speed to the robot. Among them, the tracking path is represented by the six coefficients of the quadratic equation in two variables. In the cluster control module, the host computer establishes communication with all robots in the cluster, and sends cluster control commands, robot motion status and environmental information to each robot. Cluster control commands include target position setting, target formation setting and target area setting. The target area is a polygonal area and a circular area, wherein the polygonal area is represented by the set of all vertex coordinates of the polygon, and the circular area is represented by the center coordinates and radius.

The command format sent by the host computer is: robot address + function code + data.

The advantages and positive effects of the present invention are:

(1) The swarm robot experiment platform of the present invention integrates a robot, a communication system, a positioning system and a host computer, and has the conditions for centralized and distributed swarms to complete experiments such as formation control and area coverage, which facilitates the transplantation and verification of swarm control algorithms .

(2) The swarm robot experiment platform of the present invention combines external auxiliary positioning with robot self-positioning, and the host computer can save the movement path of the swarm and monitor its running state, which is beneficial to the improvement and evaluation of the swarm control algorithm.

(3) In the swarm robot experiment platform of the present invention, the coefficients of the quadratic equation are used to describe the robot tracking path, the polygon vertex coordinate set is used to describe the target polygon area, and the center coordinates and the radius are used to describe the target circular area. Effectively reduce the data volume of the control command and reduce the communication load.

Description of drawings

Fig. 1 is a schematic diagram of a swarm robot experimental platform according to an embodiment of the present invention;

2 is a schematic diagram of the motion of a robot according to an embodiment of the present invention;

3 is a schematic diagram of a robot self-positioning using a distance sensor according to an embodiment of the present invention;

4 is a schematic diagram of functional modules implemented by a host computer according to an embodiment of the present invention;

5 is a schematic diagram of a formation of a swarm robot according to an embodiment of the present invention;

FIG. 6 is a working flow chart of the swarm robot experiment platform according to the embodiment of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

As shown in FIG. 1 , the swarm robot experiment platform according to the embodiment of the present invention includes: a host computer 1 , an experiment site 2 , a swarm robot 3 , a communication system 4 and a positioning system 5 . Among them, the swarm robot 3 is a desktop mobile robot with two-wheel differential drive, and moves in the experimental site 2; the communication system 4 is composed of multiple wireless ad hoc network communication modules, and the host computer 1 and each swarm robot 3 are equipped with A wireless communication module is installed, which can realize mutual long-distance data transmission; the positioning system 5 includes a depth camera and a positioning program. The depth camera is installed directly above the center of the experimental site 2 and is connected to the host computer 1. The positioning program calculates according to the obtained picture. The pose information of each swarm robot 3 is obtained and transmitted to the host computer 1; the host computer 1 can issue a control command of a robot or the entire swarm, and display its running status information.

The swarm robot 3 adopts a modular design program, including a decision control module, a drive module, a communication module and a sensor module, which is easy to transplant the swarm control algorithm. The decision control module can call other modules to control the movement of the robot. Before the cluster experiment, the cluster control algorithm must be written into the decision control module and programmed into the main controller of the robot. The drive module is used to drive the robot action. The communication module is a wireless communication module, and communicates with other wireless communication modules in an ad hoc network. Each robot has its own gyroscope, encoder and distance sensor. The sensing module obtains the rotation angle of the robot through the gyroscope; obtains the pulse number through the encoders of the left wheel motor and the right wheel motor to further calculate the speed of the robot; through the distance sensor, the distance to the surrounding things can be measured.

The bottom surface of the experimental site 2 is spray-painted cloth, surrounded by a fence made of KT board. The size of the site can be freely changed by lengthening or shortening the fence.

The communication system 4 is composed of a plurality of wireless communication modules based on the ZigBee protocol, which has a long transmission distance and can work stably for a long time. The address of each wireless communication module is unique. After power-on, it will automatically form a multi-hop mesh network with surrounding wireless communication modules, which can send data to any node in the network. The host computer 1 and the robot 3 connected to the wireless communication module are both nodes in the network. Based on the communication network, the communication system can realize three communication modes such as one-to-one, one-to-many, and many-to-many. One-to-one mode means that the data sent (or received) by one node (host computer or robot) can only be received (or sent) by another specific node; one-to-many mode means that data sent (or received) by one node Data can be received (or sent) by multiple nodes; the many-to-many mode means that all nodes can send data to other nodes and can also receive data sent by other nodes.

The positioning system 5 includes robot self-positioning and external auxiliary positioning, wherein the robot has two self-positioning modes.

The first self-positioning method is to use the angle and speed data output by the gyroscope and encoder carried by the robot itself to accumulate and calculate the current position of the robot, as shown in Figure 2: Set the point at the lower left corner of the experimental site 2 to As the origin, a global coordinate system is established. The pose of the robot can be represented by (x, y, θ), which are the x coordinate, y coordinate of the robot center position, and the robot rotation angle θ; its motion state can be expressed by (v, ω ) to represent, respectively, the forward speed v and the angular speed ω. The sensor module can output the pulse number n _l generated by the left wheel encoder, the pulse number n _r generated by the right wheel encoder and the robot rotation angle θ within the time Δt. When the wheel rotates once, the number of pulses generated by the encoder of the left and right motors is N, the diameter of the driving wheel is d, and the distance between the wheels is L. Let (x _i , y _i , θ _i ), (vi , ω _i ) be the pose of the robot at time _i , then there are:

v _i =(v _l +v _r )/2,ω _i =(v _r -v _l )/L

θ _i =θ,x _i =x _i-1 +v _i ·Δt·cos(θ _i ),y _i =y _i-1 +v _i ·Δt·sin(θ _i )

Among them, v _l and v _r represent the speed of the left and right wheels of the robot respectively, and the unit is mm/s. x _i-1 and y _i-1 are the coordinate positions of the robot at time i-1.

The second self-positioning method is when the robot's orientation is parallel to the X-axis or Y-axis of the global coordinate system, that is, when its angle is 0 (360°), 90°, 180° or 270°, it is measured by the distance sensor carried by itself. The distance from the experimental site wall to obtain the current position, as shown in Figure 3. The distances measured by the two distance sensors facing the left and lower walls of the experimental site are l ₁ and l ₂ , and the current position of the robot can be obtained, that is, x _i =l ₁ , y _i =l ₂ .

The accuracy of the pose calculated by the above two self-localization methods largely depends on the accuracy of the robot sensor. The noise of the encoder and the distance sensor, and the drift of the gyroscope are inevitable. A non-negligible cumulative error is gradually generated, and the initial pose of the robot cannot be obtained. Therefore, the positioning system of the present invention also provides an external auxiliary positioning method to obtain the robot pose with high precision and no accumulated error. In this positioning method, the depth camera installed above the center of the experimental site is used, and the Hough transform function is used to identify the pose of the circular robot (that is, the robot coordinate system) in the camera coordinate system, so that the robot can be calculated in the global coordinate system. pose, expressed as follows:

P ⁱ = ^j _i TP ^j

Among them, P ^j is the pose of the robot in the depth camera coordinate system, ^j _i T is the pose of the depth camera coordinate system relative to the global coordinate system, and P ⁱ is the pose of the robot in the global coordinate system.

The host computer 1 is composed of two parts, the individual control module and the cluster control module, and has the function of issuing individual or cluster control commands and displaying the status information, as shown in Figure 4. In the individual control module, the host computer can establish communication with any robot in the cluster and send its target pose, tracking path and target speed. The target pose is represented by (x _d , y _d , θ _d ); the target speed is represented by (v _d , ω _d ); the tracking path uses the six coefficients of the quadratic equation (a, b, c, d, e, f) means that the path expression is as follows:

ax ² +by ² +cxy+dx+ey+f=0

Among them, x and y are the position coordinates.

For example, a straight line path y=x can be represented by (0,0,0,-1,1,0), and a circular path with (5,5) as the center and 10 as the radius can be represented by (1,1,0,- 10,-10,-50).

The command format sent by the host computer 1 is: robot address + function code + data. After the robot receives the transmitted data, it performs corresponding operations according to its cluster control program. In addition, the host computer 1 displays the set target point and path by drawing, and updates the pose of the robot in real time.

In the cluster control module, the host computer needs to establish communication with all robots in the cluster, and transmit control commands, robot motion status and environmental information. First, you need to select the cluster communication mode, that is, centralized communication and distributed communication. Centralized communication, that is, each robot can obtain global environmental information from the host computer, including the pose, speed information of all other robots in the cluster, and its own external auxiliary positioning information; distributed communication, that is, each robot can only Obtaining local environment information within its set range from the host computer, including the pose and speed information of other robots within this range, can only rely on self-positioning, and cannot obtain its own external auxiliary positioning information. Cluster control commands include target position setting, target formation setting and target area setting. The target position is represented by (x _sd , y _sd ); the target formation includes two types of triangular formations and diamond formations, which are represented by (type, D, α), where type=0 is a triangle formation, type=1 is a diamond formation, D is the distance between the individuals in the formation, α is the angle between the individuals in the formation, as shown in Figure 5; the target area can be divided into a polygonal area and a circular area, where the polygonal area is represented by the set of all vertex coordinates of the polygon , namely {(x ₁ ,y ₁ ),(x ₂ ,y ₂ ),(x ₃ ,y ₃ ),…}, the circular area is represented by the center coordinates and radius, namely (x _o ,y _o ,r) . The status display module can display the poses of all individuals in the cluster, the target formation, and the target coverage area, which is conducive to detecting the status of the cluster and evaluating the experimental results.

In the prior art, a sequence of points located on the outline of the path and the target area with equal intervals is generally used to describe the path and the target area. The accuracy of this description method is related to the interval of the points. If it is large, the length of the point sequence will also increase, which will lead to an increase in the data amount of the control command. In the present invention, the target polygon area is described by the set of polygon vertex coordinates, and the target circular area is described by using the coordinates of the center of the circle and the radius. while reducing the amount of data.

The workflow of a swarm robot experiment platform of the present invention is shown in FIG. 6 . First, burn the cluster control program for all the robots in the cluster, put the robots into the experimental site and power on after completion; open the interface of the host computer, start the wireless communication module, and establish an ad hoc network communication network with all robots; use the external positioning system of the positioning system The module sends the positioning result to each robot through the host computer to complete the initialization of the robot pose; the host computer sends individual/cluster control commands; the robot executes the task according to the control program after receiving the command; it is updated through the self-positioning module during the execution of the task. The pose of itself is released to the host computer; the host computer displays the pose of the individual/cluster; after completing the task, the robot waits for the next command.

Except for the technical features described in the specification, they are all known technologies by those skilled in the art. The present invention omits descriptions of well-known components and well-known technologies to avoid redundant description and unnecessarily limit the present invention. The implementations described in the above embodiments do not represent all implementations consistent with the present application. On the basis of the technical solutions of the present invention, those skilled in the art can make various modifications without creative work. Or deformations are still within the protection scope of the present invention.

Claims (8)

1. a swarm robot experiment platform, is characterized in that, comprises: swarm robot, experiment site, communication system, positioning system and host computer; The swarm robot is a desktop mobile robot that adopts two-wheel differential drive and moves in the experimental site; the robot itself carries a distance sensor; the communication system is equipped with a wireless communication module on the host computer and each swarm robot, forming a Wireless ad hoc network; the positioning system includes a depth camera and a positioning program. The depth camera is installed just above the center of the experimental site and is connected to the host computer. The positioning program calculates the pose information of each cluster robot and transmits it to the host computer. The upper computer issues the control commands of a single robot or the whole group of robots, and displays the operation status of the robot. 2. The swarm robot experiment platform according to claim 1, wherein the functions in the swarm robot adopt a modular design, including a decision control module, a drive module, a communication module and a sensing module; before the cluster experiment is carried out , the cluster control algorithm is written into the decision-making control module; the driving module drives the action of the robot; the communication module is used for wireless communication; the sensing module obtains output from the gyroscope, encoder and distance sensor carried by the robot itself The data is sent to the positioning system. 3. swarm robot experimental platform according to claim 1, is characterized in that, described positioning system, the positioning program on it carries out external auxiliary positioning to robot, adopts following mode: Obtain the image captured by the depth camera, use the Hough transform function to identify the position of the robot in the camera coordinate system, and calculate the pose of the robot in the global coordinate system, as follows:

Among them, P ^j is the pose of the robot in the depth camera coordinate system,

is the pose of the depth camera coordinate system relative to the global coordinate system, and P ⁱ is the pose of the robot in the global coordinate system; the global coordinate system is established with the point at the lower left corner of the experimental site as the origin. 4. according to claim 1 or 2 or 3 described swarm robot experimental platforms, it is characterized in that, described positioning system, the positioning program on it carries out robot self-positioning, adopts following way: The point at the lower left corner of the experimental site is set as the origin, and a global coordinate system is established; the current position of the robot is obtained by accumulative calculation through the angle and speed output by the gyroscope and encoder mounted on the robot itself, including: The number of pulses generated by the encoder of the left and right driving wheel motors is N, the diameter of the driving wheel is d, and the distance between the wheels is L; let the coordinate position of the robot at the moment i-1 be x _i-1 , y _i-1 ; Obtain the number of pulses n _l generated by the left wheel encoder, the number of pulses n _r generated by the right wheel encoder and the robot rotation angle θ within the time Δt of the robot; then the pose of the robot at time i is obtained as follows:

v _i =(v _l +v _r )/2,ω _i =(v _r -v _l )/L θ _i =θ,x _i =x _i-1 +v _i ·Δt·cos(θ _i ),y _i =y _i-1 +v _i ·Δt·sin(θ _i ) Among them, v _l and v _r represent the speed of the left and right driving wheels of the robot respectively; v _i is the forward speed of the robot at time i; ω _i is the angular velocity of the robot at time i; x _i , y _i are the coordinate positions of the robot at time i, θ _i is the rotation angle of the robot at time i. 5. swarm robot experimental platform according to claim 1, is characterized in that, described positioning system, the positioning program on it carries out robot self-positioning, adopts following way: Set the point at the lower left corner of the experimental site as the origin to establish a global coordinate system; When the direction of the robot is parallel to the X-axis or Y-axis of the global coordinate system, the current position is obtained by measuring the distance from the surrounding wall of the experimental site through the distance sensor carried by the robot itself. 6. The swarm robot experiment platform according to claim 1 or 3, wherein the host computer is composed of an individual control module and a swarm control module; The host computer in the individual control module establishes communication with any robot in the cluster, and sends the target pose, tracking path and target speed to the robot; the tracking path uses the six coefficients of the quadratic equation (a, b, c, d, e, f) means that the path expression represented is as follows: ax ² +by ² +cxy+dx+ey+f=0 Among them, x and y are the position coordinates; In the cluster control module, the host computer establishes communication with all robots in the cluster, and sends cluster control commands, robot motion status and environmental information to each robot; the cluster control commands include target position setting, target formation setting and target area setting; the target area is a polygon Areas and circular areas, where the polygonal area is represented by the set of all vertex coordinates of the polygon, and the circular area is represented by the center coordinates and radius. 7. The swarm robot experiment platform according to claim 6, wherein the swarm control module is provided with two swarm communication modes, including: centralized communication and distributed communication; in the centralized communication mode, Each robot can obtain global environment information from the host computer, including the poses of all other robots in the cluster and its own external auxiliary positioning information; in the distributed communication mode, each robot obtains its set range from the host computer The local environment information within the robot, including the poses of other robots within the set range, cannot obtain its own external auxiliary positioning information. 8 . The swarm robot experimental platform according to claim 6 , wherein the command format sent by the host computer is: robot address+function code+data. 9 .

Images