Research Of Obstacle Avoidance Of Autonomous Distribution Vehicle Based On Improved Artificial Potential Field Algorithm And Grid Method

With the continuous development of driverless technology,the autonomous distribution vehicle based on driverless technology makes it possible to solve the "last kilometer" problem in logistics distribution.Autonomous Distribution Vehicle,also known as Unmanned Distribution Vehicle or Unmanned Delivery Vehicle,has the functions of autonomous positioning,navigation and real-time obstacle avoidance.It is a hotspot in the field of wheeled autonomous mobile robots.This paper mainly studies the path planning technology and environment modeling technology in the self-obstacle avoidance technology of distribution vehicles.Firstly,this paper introduces the research background,significance and research status at home and abroad of the self-owned distribution vehicle,and focuses on the analysis and summary of the research status of the distribution vehicle path planning technology and environmental modeling technology.Then,based on the kinematics model of the autonomous distribution vehicle,this paper uses the kinematics principle to analyze several typical driving motions,such as straight line,rotation and turning.Based on the sensor model,the processing principle of the perceived information in the process of autonomous obstacle avoidance of the distribution vehicle is analyzed.Finally,according to the characteristics of real-time,accuracy and efficiency of local autonomous obstacle avoidance of distribution vehicles,a hybrid autonomous obstacle avoidance scheme based on obstacle avoidance path planning technology and environmental modeling technology is proposed.In this scheme,the improved artificial potential field method is used to design the local obstacle avoidance path planning algorithm,and the improved grid method is used to design the local environment modeling scheme.When the artificial potential field method is improved,an improved scheme is adopted to supplement the energy of the target point and introduce the distance ratio function to solve the problem of target unreachability in the process of obstacle avoidance path planning,and an improved scheme is adopted to introduce virtual repulsion force to the distribution vehicle in a specific direction to solve the local minimum value when the distribution vehicle,obstacle and target point collinear in the process of obstacle avoidance path planning.In order to solvethe problem of local minimum caused by multiple obstacles in the process of obstacle avoidance path planning,an improved scheme of introducing augmentation function for gravitational potential field function is adopted,and the rationality of the improved scheme is verified by simulation.In order to improve the grid environment model,the geometric feature method and artificial potential field method are used to solve the problem of too much data in the grid environment model.Based on the improved grid environment model,the environmental modeling scheme is designed for two typical driving scenarios: local static complex obstacles and dynamic crowd.Through simulation,the feasibility and rationality of hybrid autonomous obstacle avoidance scheme based on improved artificial potential field method and grid method are verified.Compared with other obstacle avoidance schemes,this scheme is more efficient and accurate in local real-time obstacle avoidance.