Research On Path Planning Algorithm Of Mobile Robot In Dynamic And Complex Scene

As the core research content in the field of autonomous navigation,path planning is the key to reducing energy consumption,improving the safety and work efficiency of the autonomous mobile robots.The sliding differential four-wheel mobile robot is taken as the research object,and conducting research on its path planning algorithm in this paper.The main research contents are as follows:First,for the known static and complex scenes,aiming at the energy-saving planning problem with travel time,actuator force,and path smoothness as optimization indicators,a collision-constrained adaptive heterogeneous multi-objective differential evolution Lévy flight algorithm is proposed to perform global path planning.The environment is modeled by an improved vector and raster combination algorithm to reduce the number of collision detections during the subsequent path search.Hidden Markov strategy chain is used to adaptively select the heterogeneous multi-objective differential evolution mutation strategy and Lévy flight strategy.This mutation mode can maintain the balance between global search and local optimization of the population,and has better convergence performance.In order to avoid search stagnation and premature phenomena,an infeasible path correction strategy with collision constraints is defined to reduce the generation of infeasible paths and improve the algorithm's path optimization ability.The simulation results show that the proposed algorithm is more effective.Secondly,for the unknown dynamic and complex scenes,a fuzzy dynamic window approach based on the three-level buffer is proposed,which solves the problem that the dynamic window approach is difficult to carry out effective local path planning in this scene.By improving the sub-functions of the dynamic window approach and adding a cost function,the ability of the robot to travel to the target point in a complex environment is enhanced.Based on the consideration of the balance between speed and effectiveness,and considering the geometric dimension of the robot,the fuzzy controller is designed.At the same time,three safety-level buffer areas are set up for the robot,the laser radar is combined with the three-level buffer to construct the buffer factor.They are combined with the fuzzy controller,which are used to complete the control of the robot's linear velocity and angular velocity.Simulation comparison results show that the proposed algorithm significantly improves the obstacle avoidance ability of the robot in unknown dynamic and complex scenes,makes the robot drive as fast as possible.Finally,aiming at the energy-saving problem of the robots in dynamic and complex scenes,a collision-constrained adaptive heterogeneous multi-objective differential evolution Lévy flight-dynamic window approach is proposed for the hybrid planning.The algorithm enables the robot to perform local planning along the global path,taking the global path node as the sub-target point of the local path planning,and real-timely enhance the movement trend toward the sub-target point when the safety level is low.The simulation results show that the proposed algorithm can effectively avoid unknown dynamic obstacles,and quickly track the global path,which improves the working efficiency of the robot and realizes the energy saving of the system.A physical experiment platform is built to verify the effectiveness and practicability of the hybrid planning strategy through experiments.