Research On Path Planning Of Restaurant Service Robot

Path planning technology is one of the key technologies for robots to realize autonomous mobile navigation,and it is also one of the research hotspots for domestic and foreign scholars in this field.In order to improve the autonomous navigation capability of serving robots in th e restaurant environment,this paper has carried out a more recent research on the basis of a comprehensive analysis of the research status and development trend of domestic and foreign path planning technologies.The main research contents are as follows:Firstly,on the basis of traditional genetic algorithms,the static path planning of wheeled mobile robots with improved genetic algorithm(HLGA)is proposed.Using the pseudo-random Halton sequence to generate the initial path with low degree of difference as the initial population of HLGA,using the editing distance to reduce the similarity of the mutation operator in the genetic algorithm,on the Matlab,the simulation analysis of the static path planning of the restaurant environment based on the grid map,before the improvement,the HLGA planning path was safer,the distance was shortened by approximately 18.46%,the inflection point reduced by approximately 27.27%,and the efficiency increased by approximately 19.25%.Then,based on static path planning,a dynamic window algorithm based on speed avoidance(VDWA)is proposed.The speed obstacle avoidance method is used to accurately estimate the feasible obstacle avoidance area of the dynamic window,which makes the robot have higher obstacle avoidance efficiency and faster algorithm speed in the dynamic restaurant environment,which reduces the robot's runtime and memory consumption in actual path planning.Based on the simulation experiment analysis of Matlab,before the improvement of VDWA,the success rate of dynamic obstacle avoidance is increased by approximately 5.41%,the efficiency is improved by 9.08%,and the error is reduced by about 17.01%.Finally,in the actual restaurant environment,the Turtlebot2 robot platform based on the ROS development environment completed the construction of the improved algorithm and its function implementation,and analyzed the experimental results.Practice shows that the actual system built can accurately locate the robot in real time,and create a high-precision environment map at the same time.It has completed the exploration of the unknown environment and has high robustness and real-time performance.