Research On Motion Control And Path Planning Algorithm Of Wheeled Moving Robot

With the development of science and technology,robotics as a multidisciplinary field is one of the hot research fields in modern times.As China’s aging population gradually deepens,the application and popularization of moving robots in industry,medicine,agriculture,construction,military and other fields is an inevitable trend of future development.This paper focuses on the motion control and path planning problems in the key technologies of moving robots,and studies the motion control and path planning algorithms of moving robots based on the dual-wheel differential drive Tank-like chassis model,in order to realize the application of moving robots in various fields and realize industrial automation and Lay the foundation for intelligent development.This paper proposes a PID optimization strategy based on the improved differential gray wolf algorithm to solve the problem that the difficulty of PID parameter optimization leads to low path tracking accuracy of moving robots.For the global path planning algorithm and local obstacle avoidance of moving robots,this paper focuses on research and algorithm improvement on different environmental maps and obstacles.The main achievements of the paper are as follows:(1)By introducing an improved Tent chaotic map reverse learning strategy and nonlinear convergence factor,and adding mutation,crossover,and selection operations of differential evolution,this paper proposes an improved differential gray wolf optimization algorithm.The improved algorithm is applied to optimize the PID control parameters of the two-wheel differential drive Tank-like chassis model moving robot.The results of Simulink simulation and physical experiments show that the control effect of PID parameters optimized by this improved algorithm is obviously superior to other intelligent optimization algorithms,which can effectively improve the trajectory tracking performance of moving robots,so that the actual trajectory of moving robots can better fit the target trajectory.(2)A uniform random sampling algorithm is proposed to generate the initial sampling points of the traditional BIT~*,in response to the problem that the planning path obtained from the initial sampling is too long,resulting in low efficiency in the final path planning.Simulation experiments have shown that uniform random sampling can preserve randomness while possessing higher discreteness than random sampling for both complex ordered maps and complex unordered maps;Therefore,compared to RRT* and traditional BIT~*,the improved BIT~* based on uniform random sampling can improve the efficiency and stability of path planning for both complex ordered maps and complex unordered maps after multiple experiments,and find an excellent path in a shorter time.(3)By improving the evaluation function of DWA algorithm and adding the obstacle turning point evaluation subfunction,the DWA algorithm can avoid obstacles in advance when facing "C" type obstacles,so as to avoid the moving robot falling into them;Secondly,the improved BIT~* algorithm is introduced to make use of global map information,and the integrated path planning algorithm can make the moving robot have real-time dynamic obstacle avoidance ability while ensuring the global path is short.