Mobile robot path planning method in dynamic environment

Abstract

The invention discloses a mobile robot path planning method in a dynamic environment. The mobile robot path planning method specifically comprises: establishing a global grid map; screening out dynamic obstacles from the detected obstacles according to the characteristics and position change conditions of the obstacles; analyzing the motion trend of the dynamic obstacles, and determining a corresponding influence area in a set time period; for each grid in the influence region, calculating the probability of being occupied by each dynamic obstacle in the set time period, and taking the maximumprobability as the probability of being occupied by the grid; generating candidate paths from the initial position to the target position according to a given rule, and evaluating each candidate pathaccording to whether the candidate path passes through the static obstacle, the length and the probability of the dynamic grid; and taking the candidate path with the optimal evaluation result as a final planned path. According to the method, when a mobile robot moves along the finally planned path, the probability that the mobile robot collides with the dynamic obstacles is greatly reduced, andthe moving safety and reliability of the mobile robot are improved.

Description

Technical field [0001] The present invention relates to the field of robotics, in particular to a method and device for path planning of a mobile robot in a dynamic environment, a computing device and a computer-readable storage medium. Background technique [0002] During the movement of mobile robots (including self-driving cars, etc.), it is necessary to perceive, analyze and calculate environmental information, combine the mobile robot's own positioning and target position, and use obstacle avoidance and mobile robot path planning methods to achieve autonomous navigation. However, the existence of dynamic objects in the environment has caused great troubles to the navigation of mobile robots. In the process of constructing the environment map, dynamic objects may be built into the map as inherent factors in the environment, so that there may be trajectories of dynamic objects in the environment map, and mobile robots are vulnerable to their trajectories during navigation. Fa...

AI Technical Summary

Problems solved by technology

[0003] Existing obstacle avoidance methods can largely avoid the negative impact of dynamic objects on the autonomous movement of mobile robots, but they also have certain limitations: 1) The movement direction selected by the obstacle avoidance method is usually selected from a local perspective. It does not have global optimality, which makes the quality of the motion path of the mobile robot drop significantly during this period in order to avoid dynamic obstacles, such as the selected path length is longer and the smoothness is poor; 2) This type of technology usually does not consider The movement trend of dynamic obstacles is only considered from the principle of choosing the movement direction closer to the navigation target or not destroying the original planned path as much as possible, which makes it impossible to guarantee the obstacle avoidance effect, and its avoidance direction may be different from the movement of dynamic obstacles. Direction conflicts, but it is easy to collide with dynamic obstacles

Images