Detumbling Strategy For A Space Manipulator After Capturing A Target

On-orbit capture with the use of space manipulators is a premise for the implementation of on-orbit services such as on-orbit fueling,on-orbit maintenance,and clean-up of failed satellites in orbit.It is also a guarantee for further improving space exploration capabilities.In order to complete the on-orbit capture task,in addition to accurately approaching compliantly grasping the capture target,the space manipulator should also quickly and safely stabilize the combination formed by them in the postcapturing phase.Therefore,in order to improve the detumbling capability of the space manipulator in the post-capturing phase,based on trajectory planning and controller design,this paper conducts research on detumbling strategy designing,including the detumbling strategy considering obstacle avoidance,the detumbling strategy considering the flexible vibration suppression and the detumbling strategy considering the null disturbance to the base attitude.On this basis,a detumbling control system with multitask mode is established.Considering the problem that the compound system may collide with static or dynamic obstacles during the detumbling process,a detumbling strategy with obstacle avoidances called CFDS are proposed.The strategy includes trajectory planning and controller design.Firstly,the kinematic and dynamic model of the compound system is established according to its characteristic.Then,based on the principle of artificial potential field method and combined with the constraints of joint acceleration limitation,the acceleration potential field method is proposed to plan the trajectory for the compound system during the detumbling process.The planned trajectory can not only guide the movement of the manipulator to achieve the stabilization of the compound system,but also help the compound system avoid obstacles in a dynamic environment.Afterwards,the dynamic uncertainty of the compound system is considered.In order to ensure that the compound system can accurately track the above trajectory,by implementing RBFNN to compensate dynamic uncertainties,an adaptive controller is constructed.A detumbling strategy considering flexible vibration suppression is proposed.On the one hand,considering the problem of the residual vibration of the flexible attachment on the base,a detumbling motion planning algorithm is proposed.First,a comprehensive evaluation index is designed according to the requirements of avoiding self-collision,avoiding singularity and improving path smoothness.Then,combined with the idea of the optimal rapid random tree algorithm(RRT*),and using the above comprehensive evaluation index to design the cost function,the path planning of the compound system is completed.On this basis,an optimization model with the trajectory of the compound system as the variable and the minimum residual vibration of the flexible attachment as the goal is constructed,and a PSO algorithm is developed to solve the optimization,so as to obtain the detumbling trajectory with minimal residual flexible vibration.On the other hand,a backstepping adaptive controller is constructed to solve the precise control problem of a flexible-joint space manipulator.The backstepping design process is based on a modified state space equation.An extended-state observer is included in the controller,which realizes real-time estimation of joint speed,link speed and joint stiffness uncertainties.An RBFNN compensation term is designed in the controller to eliminate the impact of dynamic uncertainties.Taking the free-floating space manipulator as object,a detumbling strategy considering the null disturbance to the base attitude is proposed.Based on the principle of momentum conservation,the influence of the motion of the manipulator and the grasped target on the attitude of the free-floating base is analyzed,and the dynamic model of the attitude disturbance of the base is deduced.On the basis of this model,combined with the Lyapunov principle,a base reactionless controller is designed to realize the nulldisturbance to the base attitude at the displacement level.Considering that the dynamic uncertainty of capturing the target will weaken the control accuracy,a compensation term is designed in the base reactionless controller,so that an adaptive base reactionless controller is formed.In order to realize the compound system stabilization,a detumbling control vector is added in the adaptive base undisturbed controller.The control vector is designed in the null space of the base attitude.In order to obtain the detumbling control vector,an optimization model with the minimum joint velocity as the goal and the joint acceleration limit as the constraint condition is constructed,and the optimal result is solved by means of recursive least squares(RLS).The experiments of detumbling mission under single constraints and multiple constraints are carried out.On the one hand,according to the different characteristics of the above three detumbling strategies,three ground semi-physical experimental systems are designed.Aiming at the performance verification of the above three complex stabilization strategies,detumbling experiments under a single constraint are carried out based on the experimental systems.On the other hand,by integrating the proposed detumbling strategies,a detumbling control system with multi-task mode is proposed.On this basis,detumbling experiments under multiple constraints are carried out.The experimental results not only demonstrate that the proposed detumbling strategies have good performance in the detumbling mission under a single constraint,but also reflect that the strategies can be effectively integrated to complete the detumbling mission under multiple constraints.