Adaptive dynamic window unmanned vehicle real-time obstacle avoidance method based on danger coefficient

Abstract

The invention relates to an adaptive dynamic window unmanned vehicle real-time obstacle avoidance method based on a danger coefficient. The method comprises the steps of initializing unmanned vehicle state information; obtaining state information of the dynamic obstacle through a sensor; calculating a reachable dynamic velocity vector window according to the sampled linear velocity and angular velocity; establishing a danger coefficient of the dynamic obstacle according to the obtained state information; judging whether to start obstacle avoidance according to the danger coefficient; if obstacle avoidance is needed, the evaluation parameters are adaptively adjusted according to the danger index of the dynamic obstacle; in the reachable velocity vector window, selecting an optimal obstacle avoidance velocity vector according to an evaluation function; outputting the magnitude and direction of the optimal obstacle avoidance speed; and repeating the cycle until the unmanned vehicle avoids the dynamic obstacle and reaches the target. According to the invention, by establishing the danger coefficient based on the motion state of the dynamic obstacle and dynamically adjusting the evaluation parameter of the dynamic window method in real time based on the danger coefficient, real-time obstacle avoidance of the unmanned vehicle in a dynamic environment is realized.

Description

technical field [0001] The invention belongs to the field of real-time obstacle avoidance of unmanned vehicles, and in particular relates to a real-time obstacle avoidance method of an self-adaptive dynamic window unmanned vehicle based on a risk coefficient. Background technique [0002] Real-time obstacle avoidance technology is an important field of self-driving unmanned vehicle research. It is the key technology to realize autonomous navigation and obstacle avoidance of unmanned vehicles. Its main task is to enable unmanned vehicles to quickly avoid moving obstacles in a dynamic environment. In order to ensure the efficiency and safety of unmanned vehicles, the real-time obstacle avoidance algorithm is the core algorithm of unmanned vehicle dynamic obstacle avoidance, and it is the key to reflect the intelligence level of unmanned vehicles. [0003] In the real-time obstacle avoidance algorithm of unmanned vehicles, the dynamic window method is a classic real-time obstac...

AI Technical Summary

Problems solved by technology

However, although the dynamic window method can achieve a good obstacle avoidance function for static obstacles, the obstacle avoidance ability of this method is obviously insufficient in the dynamic environment. The reason is that the evaluation function coefficient of the dynamic window method is fixed in the planning process. The evaluation coefficient cannot be adaptively adjusted for obstacles in different states to adjust the real-time obstacle avoidance ability of unmanned vehicles

Images