Multi-robot formation positioning method based on particle filter and robot equipment

Abstract

The invention discloses a multi-robot formation positioning method based on a particle filter and robot equipment, and relates to the technical field of robots and navigation. The method comprises thefollowing steps: 1, establishing a grid map; 2, allowing multiple robots to realize cooperative formation motion; 3, iteratively updating the particle filter; 4, allowing the robots to acquire the particle average weight of the adjacent robots in the formation in real time; 5, allowing the robots to correct the average weight of the robots in real time by utilizing the average weight of the particles of the adjacent robots; and 6, repeating the step 3, outputting the poses of the robots in real time, and stopping iterative updating until an iterative termination condition is met. The equipment comprises a grid map establishing module, a formation communication module, a formation control module, a particle filter module and a correction module. According to the invention, the positions ofthe robots can be quickly and accurately positioned at any position in navigation, a lost robot can be corrected, and the more accurate formation control can be realized based on the positioning method.

Description

technical field [0001] The invention relates to the technical field of robots and navigation, in particular to a particle filter-based multi-robot formation positioning method and robot equipment. Background technique [0002] With the improvement of embedded computing and communication capabilities, as well as the development of distributed or non-centralized ideas, it is increasingly recognized that multi-agent systems can complete more complex tasks at a lower cost than single agents. , the application requirements of multi-robot formation are increasing, especially the autonomous navigation function of robot formation. The automatic navigation system must first load the grid map generated by the mapping system, and realize the robot's self-positioning and navigation on the map. Therefore, no matter what kind of environment it is in, the robot needs to locate itself very accurately in order for the automatic navigation system to take effect quickly. [0003] The self-lo...

AI Technical Summary

Problems solved by technology

[0004] A single robot only relies on its own odometer information and radar information for particle filtering, and cannot effectively correct its own pose;

[0005] In the prior art, CN 201811205306.5 (a multi-robot collaborative positioning and control method) provides a positioning method based on multi-robot collaborative control, which is based on the distance measurement information obtained from the world coordinate system coordinates and sensors in several robots The robot position is optimized through the gradient descent algorithm. This method requires the robot to obtain a lot of data, the condition requirements are very strong, and a large amount of global calculations are required.

[0006] Therefore, those skilled in the art are committed to developing a particle filter-based multi-robot formation positioning method and robotic equipment to solve the positioning errors caused by factors such as single robot odometer errors and inaccurate laser radar scanning in the prior art. Accuracy and under-calibration problems

Images