Research On Modeling,Motion Control And Path Point Visual Inspection Of SCARA Rotary Drilling Manipulator

The construction of urban water supply pipelines is a common task in urban construction.Excavation of foundation pits is the primary link of construction,and its safety and efficiency are of great significance to the success of the entire project.At present,the project adopts manual control as the main method and machinery as the auxiliary method.Before each operation,it is necessary to formulate a detailed construction plan and conduct safety training,which consumes a lot of manpower and material resources.Considering the above problems,this dissertation studies the SCARA rotary excavation manipulator used in excavation scenarios,and proposes its motion control method and path point visual detection algorithm,which will help to improve the degree of automation of the rotary digging system and reduce labor costs.This dissertation mainly conducts research from three aspects:(1)Establish the model of SCARA rotary digging manipulator,and conduct specific research on its trajectory planning and trajectory tracking strategy.According to the operation plan of the manipulator in the excavation process,the five-degree polynomial interpolation method of joint space,the cubic B-spline curve planning method that introduces the compensation term,and the linear interpolation strategy based on sinusoidal acceleration is respectively proposed.In order to improve the operating efficiency,the fitness function is designed in combination with the operating speed,joint torque and other constraints.The trajectory running time is taken as the objective function.The optimal solution is obtained through genetic algorithms with improved crossover rate and mutation rate.Then the optimal time trajectory can be found.The joint controller of the nominal model is designed using the calculated moment method,and we design the synovial controller based on the improved reaching law to approximate the uncertain part of the system.The two parts constitute the main controller of the system.The controller receives the deviation information of speed and acceleration,and outputs the corresponding control quantity.The simulation experiment results show that the planning strategy can ensure the continuity of the trajectory,and the designed controller can ensure that the response time of the manipulator and the trajectory tracking error are within the required range.(2)Use the ellipse detection algorithm based on supporting arc to quickly identify the center position of the laser spot,and use it as a path point to provide path point input information for the calculation of the desired trajectory.Drawing lessons from the idea of LSD algorithm,by limiting the level-line angle interval of adjacent edge points,the algorithm only retains arc information.Through the verification of the effectiveness of arc segments,the effective arcs are grouped and paired,the initial ellipse set is obtained by fitting,and after clustering and deduplication,the final result is obtained through the goodness index of the fitting and the comparison of the target color information.When the initial ellipse set is generated,the completeness index is used to select the effective arc group with greater completeness;in the process of determining the support points,the search area is limited to the specified rectangular area to avoid the full-image search and accelerate the algorithm operation.After experimental analysis,the detection accuracy of the center position is about 0.6 pixels,and the response time of a single picture is about 0.5s,which meets the application requirements.(3)Design the SCARA rotary digging manipulator system and verify its main functions.Combined with the application scenarios of the SCARA rotary digging manipulator,the system's hierarchical structure and function realization methods are analyzed from the hardware and software levels.For the camera calibration in the sensing system and the task scheduling strategy in multi-mode,specific research and realization are carried out.Based on the built verification platform,camera calibration,trajectory planning and trajectory tracking experiments have been carried out successively to verify the positioning performance of the system and the effectiveness of the control algorithm.