Research on Improved Obstacle Avoidance Algorithm

The traditional artificial potential field algorithm has the shortcomings of appearing trap area easily and shaking in front of obstacles. This paper, studying traditional artificial potential field deeply, optimizes the sensor data with the introduction of the BP network and the GA algorithm, unmanned vehicles achieving autonomous obstacle avoidance and prognoses in advance when passing the path with obstacles, and effectively eliminating system instability caused by shock and dead zone. Comparing with traditional artificial potential field algorithm, the improved algorithm makes a further improvement in the respects of system stability and efficiency.