Research On Obstacle Avoidance Of Warehousing Logistics AGV

Effective and low-cost obstacle avoidance function is an important indicator to assess the safety and economy of warehousing logistics AGV.A novel algorithm and the corresponding hardware are proposed to provide a low-cost obstacle-recognition and obstacle-avoidance solution for the warehousing logistics AGV.The specific contents are as follows:(1)According to the characteristics of the storage environment,the driving mode of the storage logistics AGV is determined.The selection of the motor and the driver module is based on the requirements of the acceleration time,maximum speed,maximum load and other requirements of the warehouse logistics AGV,and the selection of the battery is determined by the battery life requirements of the warehouse logistics AGV.Based on the obstacle avoidance requirements of the warehouse logistics AGV,the sensor scheme is determined and the working principle of the sensor is analyzed.(2)Combining the low-cost requirements of warehousing and logistics,a hardware solution of a dual-sensor fusion sensing system of "ultrasonic + laser" was designed,and an obstacle contour recognition strategy based on the nearest bump position adaptively adjusting the scanning angle was proposed.The hardware cost is only 20% of the lidar sensor system which realizes the complementary advantages of sensors,and the contour recognition function of the lidar is realized through obstacle recognition strategies and methods.First,the layout of the sensing system was determined according to the range of detection in front of the AGV,and the detection area of the ultrasonic ranging system was analyzed discretely.Secondly,the location of the nearest bump on the obstacle is estimated through the ultrasonic ranging system,and the scanning angle required by the laser sensor is calculated through the position of the nearest bump.Finally,the scanning of the laser sensor is combined with the motion path of the AGV to obtain the distance of a set of obstacle contour points,and the calculated obstacle contour point coordinates are processed.In addition,this paper uses linear regression method to calibrate the three sets of ultrasonic sensors in the ultrasonic ranging system.(3)Aiming at the problem that the original path will be changed by the obstacle avoidance programming method of "planning first and smoothing later".the obstacle avoidance programming problem is transformed into an optimal control problem to solve,and the motion trajectory is optimized and smoothed without changing the original path.Firstly,the motion constraints,collision avoidance constraints,path channel constraints and two-point boundary conditions of the warehouse logistics AGV are quantified,the constraints of the optimization problem are established,and the objective function is established with the shortest obstacle avoidance time as the goal.Secondly,The optimal control problem is discretized into a constrained nonlinear programming problem based on the direct collocation method,and the non-linear programming problem is optimized by direct collocation method.Finally,calculation examples are used to verify the optimized and smoothed results of motion trajectory.The research results show that in the environment of moving or static obstacles,the direct collocation method can be used to plan a trajectory that meets the motion constraints of AGV.(4)The method of obstacle recognition and avoidance in this article is verified by setting up an experimental environment.About the position recognition accuracy of the nearest bump on the obstacle,The evaluation index is established by the average error of the horizontal and vertical coordinates.In terms of the accuracy of obstacle contour recognition,the accuracy evaluation index of contour recognition is established by the absolute value of the difference between the size or position identification value and the corresponding actual value.In terms of trajectory planning and smoothness of obstacle avoidance movement,an obstacle avoidance trajectory can be successfully planned in single or multiple static obstacle environments based on the direct collocation method,and the trajectory is relatively smooth,satisfying the motion constraints of AGV.The warehousing logistics AGV obstacle recognition and obstacle avoidance software and hardware scheme proposed in this paper can not only meet the requirements of low-cost design,but also realize the accurate identification of the contour and position of obstacles,as well as the optimization and smoothing of the obstacle avoidance movement trajectory under the condition of dynamic and static obstacles.