Obstacle Avoidance Path Generation Algorithm for Quadruped Inspection Robot in Cable Tunnel

When modeling and analyzing the static environment of cable tunnels, dynamic obstacles cannot be processed in real time. Therefore, a four legged inspection robot obstacle avoidance path generation algorithm for cable tunnels is proposed. By calculating the distance between the ultrasonic receiver and transmitter for localization, the inertia weights of different ppapers are dynamically given based on the current overall ppaper condition. The ROS in the robot control module completes global obstacle avoidance path planning using the A * algorithm, and uses the dynamic window method to complete local obstacle avoidance path planning, automatically generating the optimal obstacle avoidance path for the robot. The experimental results show that the obstacle avoidance path generated by the proposed method has a short distance and can effectively avoid all obstacles. After 30 attempts, the collision rate is 3.02%, and the root mean square error of the robot's actual trajectory distance from the set path is 0.0993 m.