Optimal Control For Space Manipulator Based On Motion Reliability

Space manipulator as a kind of complex electromechanical system is playing more and more important role in the development of aerospace industry and space exploration. The motion reliability of space manipulator has significant influence on task completion on-orbit during its long lifetime service. When operating on-orbit, space manipulator is affected by radiation,high-low temperature alternating of space environment, many kinds of malfunctions and failures are inevitable. Since it is of great cost to achieve failure components maintenance and replacement on orbit, it has great significance in improving the ability of task completion and reducing the impact on healthy components caused by failures. Further, stable and safe operations of space manipulator should be achieved during long lifetime service on-orbit. For this purpose, motion reliability control strategy is researched in the paper, which aims at achieving motion performance optimization when space manipulator is healthy and maximizing the probability of task completion when failure occurs. The research conforms to the goals of stable and safe operations in the space manipulator long-term development strategy, which also has significant promotion in economic and practical application.This research of the paper is supported by the National Key Basic Research Program of China "Basic theory research on operational reliability system control of space mechanism"(2013CB733005). Space manipulator is treated as research object and motion reliability is treated as research goals.Technologies related to motion reliability evaluation, motion reliability optimization control for healthy and failure state of space manipulator are analyzed in the paper, which can be detailed described as follow:1. Motion reliability evaluation model establishment of space manipulator. Aiming at typical tasks on-orbit, the mathematical expression of motion reliability is derived. Then considering the influence of factors having effect on position accuracy of space manipulator, multi-layer mapping model is concluded. As a basis, the evaluation model for motion reliability is established. In order to achieve model solution, the equivalent extreme value principle is applied to simplify the model. Then the principle of maximum entropy is introduced and the probability density function can be calculated to achieve motion reliability evaluation for typical on-orbit task of space manipulator.2. Space manipulator optimization control based on motion reliability.The factors having effect on motion reliability is analyzed, and sensitivity analysis method based on nonlinear response surface method is proposed by considering on-orbit task constraints. As a result, the sensitivity of factors can be calculated fast. According to the sensitivity analysis results, factors with high sensitivity are chosen as decision variables. Then Considering the constraints of configuration, environment and tasks, a presentation of motion reliability is proposed based on multiple sub-objects. Further, optimization control model for motion reliability is established based on multi-indexes and multi-factors. In order to achieve model calculation, covariance matrix is introduced to convert multiple objectives into single objectives for simple calculation.3. Fault tolerant evaluation and strategy design for space manipulator joint failure. A motion performance evaluation system is designed for space manipulator joint-locked failure. Since the traditional fault-tolerant space is so small for target position arrangement, reliable fault-tolerant space is proposed by considering joint reliability. As a result, the feasible region of target position is expanded efficiently. Moreover, based on motion performance evaluation and operational redundancy analysis of on-orbit tasks,fault tolerant strategy is designed which treating the motion performance evaluation as input and treating fault tolerant methods for different operating conditions as output when joint failure occurs.4. Reduction methods for parameters jumps caused by joint-locked failure. Aiming at task completion and motion performance optimization, the reduction methods for parameters (velocity and torque) jumps are discussed.Firstly, the models of velocity jumps and impact torque are established. By taking the gradient of manipulability as parameters' compensation weight, a global compensation term is constructed and the parameters are compensated during the entire task cycle. With global compensation, the time delay problem can be avoided. Then the minimization of parameters jumps is treated as objective function and the coefficient of compensation term is treated as decision variables. With optimization method, parameters jumps can be significantly reduced for arbitrary single joint failure.5. Motion reliability optimization control for typical tasks at state of joint-locked failure. For no-load operations and load operations of space manipulator, the models of. motion reliability optimization control are established. Aiming at different task constraints and different parameters jumps, reduction strategy for multiple parameters jumps are analyzed.Considering the unexpectedness characteristics of joint-locked failure, an optimal coefficient function fitting method is proposed based on prior failure information. Then the parameters reduction can be achieved no matter when joint-locked failure occurs.6. Experiments on motion reliability optimization control for space manipulator. In order to guarantee the requirement of validating theories related to motion reliability optimization control of space manipulator,sensors placement and data collection system is designed on the flotation and microgravity experiment platform, and the human-computer interaction software is improved. Then experiments for on-load operation and load operation is designed. With experiment results, the feasibility and efficiency of theories related to motion reliability optimization control are verified.