Rsearch On Walking Trajectory Control Of Tracked Hydraulic Excavator Based On SLAM And IMU Fusion Positioning

Excavator is a common heavy construction machinery,which plays an extremely important role in the field of infrastructure such as construction.Taking into account factors such as the bad working scene of the excavator and the high risk factor,the demand for automation and autonomous operation of the excavator is increasing day by day.Many scientific research institutions and enterprises have also done a lot of research work on the intelligent operation of excavators.The positioning of the excavator and the control of the trajectory of the moving mechanism of the excavator are the basis for the automation and autonomous operation of the excavator.This article focuses on the precise positioning of crawler hydraulic excavators and the precise control of walking trajectories.First,a theoretical analysis of the hydraulic system related to the excavator drop-off mechanism and the dynamics and kinematics of the excavator while walking is carried out.Then based on the binocular camera and IMU sensor data fusion,to achieve more accurate positioning of the excavator.Finally,based on the fusion positioning results of the binocular camera and the IMU,the slip parameters in the movement process were identified,a trajectory tracking controller with slip compensation was established,and an experimental analysis was performed to achieve precise control of the excavator’s walking trajectory.The main contents of this paper is organized as follows:(1)The hydraulic system is modeled on the walking device.The whole hydraulic system model is divided into electro-hydraulic proportional pressure relief valve model,three-position six-way open center proportional directional valve model,and valve-controlled hydraulic motor system model.Considering the influence of track slip factors,the kinematics and dynamics models of the excavator during the movement were established.(2)The error analysis of IMU equipment is carried out,and the IMU navigation and positioning model is established;the camera imaging model,calibration principle,commonly used feature extraction algorithm,optical flow tracking algorithm are theoretically analyzed and introduced;based on the common optical flow tracking algorithm In the above,the image pyramid is introduced to meet the needs of large-scale motion;at the same time,based on NVIDIA’s CUDA model,the GPU is used to accelerate the optical flow tracking algorithm;on this basis,a binocular positioning model is established.Finally,IMU’s inertial navigation positioning experiment,binocular camera calibration experiment,and binocular camera positioning experiment were conducted separately;the advantages and disadvantages of binocular vision SLAM(Simultaneous Localization and Mapping)positioning and inertial navigation positioning were analyzed through experiments,and the demonstration was demonstrated.The necessity of sensor data fusion positioning.(3)Fusion of sensor information of binocular camera and IMU.The tight coupling technology based on image feature points and IMU measurement information is studied,and various measurement noises and measurement deviation values of accelerometers and angular velocity meters are fully considered.To prevent the repeated integration of IMU measurement information in the back-end nonlinear optimization process,the IMU predictive sub-technology was studied to improve the calculation efficiency.Finally,a single linear positioning experiment and a right-angle turning positioning experiment were carried out in a good environment,and a closed area positioning experiment was carried out under harsh environmental conditions;the experimental results show that the positioning accuracy of the fusion positioning device is about 1.4%,which basically meets the excavator Positioning needs.(4)Based on the fusion positioning results,the sliding parameter identification model of the tracked vehicle during walking is designed,and an extended Kalman filter is designed to observe the sliding parameters during walking.Based on the identification results of slip parameters,a trajectory tracking controller with slip compensation is designed.To quickly track the target speed and angular velocity output by the trajectory tracking controller,a fuzzy adaptive PID controller is designed.Finally,using a trajectory tracking controller with slip compensation and a common trajectory tracking controller,a speed control experiment based on fuzzy adaptive PID,a slip parameter identification experiment based on general kinematics and a trajectory based on slip compensation The tracking experiment proves the effectiveness and superiority of the slip compensation through the comparison of the experimental results of the two controllers.