Simulation and verification for obstacle avoidance path tracking of unmanned agricultural machinery

Objective In order to facilitate the verification of the effectiveness of obstacle avoidance path tracking for unmanned agricultural machinery, reduce the number of actual unmanned agricultural machine tests and consumables, a simulation verification method for obstacle avoidance path tracking for unmanned agricultural machinery was designed, and a simulation verification application system was constructed.

Method Based on unmanned agricultural machinery, an integrated simulation and verification application system was constructed by integrating real complex terrain environment simulation, actual operation agricultural machinery simulation, and path planning algorithm implantation. Based on three-dimensional SLAM technology, environmental point cloud data was collected to achieve simulation modeling of farmland terrain environment. A simulation modeling of agricultural machinery was completed using unmanned agricultural machinery combined with ackermann steering mechanical structure. A TEB local path planning algorithm based on agricultural machinery dynamics constraints was proposed. Path planning tracking and obstacle avoidance were implemented in the simulation verification application system, and the effectiveness of the algorithm was verified through multiple tests.

Result The comparison test of obstacle avoidance path tracking effectiveness and the verification test of obstacle avoidance path tracking smoothness showed that the unmanned agricultural machine could effectively avoid obstacles dynamically during driving, with the shortest effective obstacle avoidance distance of 4.1 m. The path tracking control effect was good. When the distance between obstacles was greater than 5.0 m, the controllable average error was within 0.430 5 m, and the root mean square error was within 0.315 1 m. When the distance between obstacles was 4.5−5.0 m, the controllable average error was within 1.353 8 m, and the root mean square error was within 1.612 6 m.

Conclusion The improved TEB algorithm proposed in this article has strong operational capability and high operational accuracy, which meets the needs of simulation verification for obstacle avoidance path tracking in agricultural machinery navigation. This algorithm can be applied to obstacle avoidance path tracking of unmanned agricultural machinery in actual farmland environments in the future. This application system is easy to expand and can provide a foundation for the optimization design research of agricultural machinery operation status in precision agriculture for various complex working environments.