The Navigation Based on Hybrid A Star and TEB Algorithm Implemented in Obstacles Avoidance

摘要

Taking driverless and autonomous navigation as an example, artificial intelligence has gradually entered people's vision, and applications such as sweeping robots, rescue robots, safety inspection robots, and driverless cars are playing an increasingly important role in people's daily life. In this paper, two-dimensional Lidar, depth camera, odometer system, etc. are used as the sensor input data of the robot. AMCL (Adaptive Monte Carlo Localization) method is adopted to establish the autonomous positioning function of the robot. Hybrid A Star global path planning algorithm and TEB (Time Elastic Band) local path planning algorithm are used to realize the navigation function of the robot. For the indoor environment, this paper mainly selects the laboratory environment, sets different starting points and ending points, and evaluates the path planning of the robot under static obstacles and dynamic obstacles. The navigation function of the robot is tested according to two important indexes: whether collision occurs during the actual movement of the robot and whether dynamic obstacles can be avoided. The experimental results show that the robot can avoid the static obstacle and the dynamic obstacle well under the above scheme.