Research On Space Robot Trajectory Planning Method Using Monocular Cooperative Target Visual Information

As human beings become more and more interested in exploring the unknown space of the universe,space telescopes are particularly important as the main observation tool.However,with the continuous improvement of the requirements for the observation accuracy of the telescope,the diameter of the telescope has continued to increase,which has surpassed the carrying capacity of the existing launch vehicle.Therefore,the robot assembly of the super-large-aperture space telescope in orbit has become a new research direction.When the robot performs the on-orbit assembly task,it involves the grasping of the telescope sub-mirror module.In order to enable the robot to complete the work safely and efficiently,this article mainly focuses on the target object's pose measurement and robot trajectory planning.First,this article introduces the basic theory of camera imaging,derives the camera internal matrix and distortion coefficients,mean while completes the internal parameter calibration of the camera that used in the subject.A detailed derivation of the external parameter calibration algorithm of the hand-eye camera in Eye in hand mode has been done,and a simulation experiment has been done on the hand-eye calibration algorithm.The results show that the algorithm meets the accuracy requirements.Secondly,according to the background of the subject,a cooperative target that meets the requirements and is easy to detect is designed,and a new circle/ellipse feature detection algorithm is proposed for the characteristics of the target.The algorithm is mainly composed of three parts: image preprocessing,preliminary detection of circle/ellipse features,and precise detection of circle/ellipse features.To test the actual performance of new circle/ellipse feature detection algorithm,a feature detection experiment was carried out using simulated images and actual target images.The experimental results show that the circle/ellipse feature detection algorithm has relatively good detection capabilities.The principle of the EPNP relative pose calculation algorithm is introduced in detail.The accuracy measurement experiment of the vision measurement system is carried out.The results prove that the vision measurement system designed in this paper has a good measurement accuracy.Then,the kinematics analysis of the seven-degree-of-freedom redundant robot used in this paper is done.The common methods for describing robot pose information are introduced.The D-H coordinate system is built for the robot.Based on the D-H parameters,the forward kinematics solution of the robot is analyzed,and the inverse kinematics solution of the redundant robot using joint angle parameterization is derived in detail.,explains how to select the optimal solution for inverse kinematics.Finally,this paper introduces the general methods of robot trajectory planning.A 3-5-3 polynomial time-impact optimal trajectory optimization method based on PSO is proposed for optimizing joint space trajectory performance.Under the premise,the time consumed by the robot movement can be reduced,and the impact during the movement can be reduced.The robot trajectory planning joint simulation platform is built,and the trajectory optimization algorithm is tested.The results show that the algorithm has an obvious optimization effect on the robot trajectory.