Design And Implementation Of Multi-Robot Path Planning Based On Model Predictive Contouring Control

In recent years,autonomous mobile robots have played an important role in manufacturing,e-commerce,and defense technology.Multi-autonomous mobile robot systems can cooperate with each other to efficiently complete tasks to improve work efficiency.Multi-robot path planning has become a hot topic in the field of robotics.However,the current multi-robot path planning algorithms have problems such as poor real-time performance and poor coordination between robots.In this context,this thesis studies the local path planning problem of multi-autonomous mobile robot systems.In this thesis,a grid map is used to construct a map of the robot’s workspace,and a distributed model predictive contouring control(MPCC)model is established.The robot’s path planning and tracking control are combined through a constraint optimization method.This method does not need to design additional tracking controllers and can be quickly deployed on practical robots.In addition,the kinematic constraints of the robot are fully considered,and the collision avoidance requirements of the robot for the static environment and dynamic obstacles are converted into constraints for explicit processing.The contour error and the time for the robot to complete the task are two opposing control objectives.A balance of these two control objectives is achieved by minimizing the cost function in a limited time domain.The robots exchange information in a parallel manner,which ensures that the robots can solve their respective optimization problems independently and improves the scalability of the entire system.Then,the open-source Robot Operating System(ROS)is used as a platform to build the simulation environment and implement the code of the algorithm,and the optimization problem is solved quickly with the help of ACADO toolkit to provide real-time guarantee for coordinated collision avoidance among multiple robots.The classical dynamic window method,the priority-based model predictive contouring control method and the distributed model predictive contouring control method designed in this thesis are used to conduct simulation tests using different numbers of robots in several different working environments,and to compare and analyze the time to complete the task,the distance traveled and the robot’s trajectory.The simulation results show that the time to complete the task and the distance traveled by the path planning method designed in this thesis are the shortest among the three methods,which verifies the effectiveness and superiority of this path planning algorithm.Finally,the above research is used as the theoretical basis to build a physical verification scenario of the same scale as the simulation environment,using the laboratory Rikirobot robot.The experimental results show that the robots can effectively achieve collaborative collision avoidance among themselves to complete the specified tasks,which verifies the feasibility of the multi-robot path planning method in this thesis,and has certain theoretical guidance and practical reference significance for the practical production application of multi-robot systems.