Adaptive Obstacle Avoidance Optimization Algorithm Based on Learning from Demonstration

The traditional adaptive obstacle avoidance algorithm based on learning from demonstration has two problems. First, it only relies on the robot end data for obstacle avoidance without considering the linkage-based obstacle avoidance, which leads to a low success rate of robot obstacle avoidance. Second, it does not perform invalid data rejection before reorganizing data, which leads to low computational efficiency. This paper proposes an optimization algorithm of adaptive obstacle avoidance based on learning from demonstration. Which can solve the above problems, increase the success rate of robot obstacle avoidance and enhance computational efficiency.