Research On Efficient And Stable Motion Trajectory Planning For Unmanned Excavators

Along with the expansion of AI and other technologies,the development of unmanned excavators is gradually taking place.Trajectory planning,as the basis for the development of unmanned excavators,can provide a feasible reference trajectory for trajectory tracking so that the task is completed efficiently.This paper takes the SWE50 E excavator as the research object and carries out research on efficient and stable trajectory planning technology for excavation operations based on convex optimization theory.As the environment and load change continuously,the driver needs to adjust the handle in time according to the on-site construction situation during the operation.Compared with the planning method based on conventional industrial robotic arms,the method of imitating driver skills is more suitable for excavator operation trajectory planning.In this paper,driver skill information is collected using sensors and processed using data alignment,filtering,and averaging.The DPA is used to sparse the paths,laying the foundation for subsequent research.In addition,the transformation relationships between the three types of spaces are also analyzed to provide the appropriate theory for subsequent validation.Efficient operation is a constant goal for unmanned excavators.Traditional methods have yet to fully utilize the excavator’s motion performance,resulting in low efficiency.This paper proposes an efficient motion trajectory planning method.Two functions are introduced to convert the non-convex problem into a convex one,and the original problem is reconstructed into a second-order cone planning form after discretization.The experimental results show that the method can maintain a fast action state and effectively improve the operational efficiency of the excavator.Along with the fast motion of the excavator,the tips of the bucket teeth can vibrate a lot,causing the excavator to move unsteadily.This result will increase the trajectory tracking error and cause it to travel redundant paths.This paper proposes a stable motion trajectory planning method.The excavator motion smoothness problem is formulated as a convex quadratic programming problem and implemented based on an approximately optimal time allocation scheme.Experimental results show that the method can effectively reduce bucket tooth tip vibration and improve the smoothness of excavator motion.Unmanned excavators need to be efficient and smooth in motion,but efficiency is usually the main focus.Traditional methods require manually assigning appropriate weights to the objective function and reducing the optimization space during problem construction.A multivariate coupled optimization method is proposed in this paper.In studying the motionsmoothing problem,only linear equation constraints are considered to ensure that a feasible solution can be found for any time allocation.The time interval is taken as the optimization variable,and an inequality constraint is introduced to generate the time-optimal trajectory.Experimental results show that the method outperforms other methods in the combined evaluation of motion smoothness and efficiency.