Excavator Motion Planning Based on Adaptive Genetic Algorithm and Improved R-MAP

Traditional excavators operate in harsh environments and rely on driver experience, resulting in problems such as low efficiency and human error. This study proposes a new task-oriented continuous excavation action planning model. An adaptive genetic algorithm is applied on the improved R-MAP constructed by the excavator kinematic model to plan the excavator action path to improve excavation efficiency and accuracy and reduce operating costs and risks. This paper considers the constraint relationship between the bucket heading angle and the tooth tip moving direction and the possible “dead point” problem to improve the traditional R-MAP, and designs a genetic algorithm optimization framework suitable for excavator working device action planning, including path initialization coding scheme, Fitness function and genetic operator design. Combined with the improved R-MAP, the adaptive genetic algorithm can effectively search the action path of the excavator's working device within the reachable area of the working device, improving the efficiency and practicality of excavation action planning. The simulation results show that the method proposed in this study can plan all trajectories for a given excavation terrain task and control the error within 3%. Finally, the model experimental machine carried out full-shake excavation according to the planned path, completed the excavation task, and verified the effectiveness of this method.