In this paper, the time-optimal path tracking problem of the dual-arm free-floating space manipulator (FFSM) system is solved by using convex programming. Firstly, we construct a continuous nonlinear optimization problem that considers system dynamics and constraints along the path, including torque, velocity and acceleration. Subsequently, the original optimization problem is transformed into a convex optimal control problem by using scalar path coordinate and nonlinear variables. To characterize the problem in a more manageable form, we apply the direct transcription method and discretize the problem into a sparse convex optimization program. Numerical results demonstrate that the proposed approach efficiently determines the trajectory of the FFSM along the preset path. Meanwhile, the base's velocity can be limited to a small region, which is an essential requirement for ensuring the stability of the satellite. Furthermore, an experiment is conducted using GLUON robots. The tracking error between the actual and preset pose of the target verifies the feasibility of the proposed method.

A free-floating space robot equipped with multiple reconfigurable manipulators is designed and investigated in this paper. Lockable passive cylindrical joints (PCJs) are utilized to make the manipulator have the ability of changing its length and twisted angle. Each cylindrical joint, connecting two adjacent rigid links, has no embedded actuators but a brake mechanism. Normally, the mechanism is locked during the operation. When in the reconfiguration stage, two manipulators grasp each other to form a closed loop. Then one PCJ is unlocked, whose relative rotation and translation can be changed by the active torques at other joints. This system is a typical space multibody system. The dynamics of the space robot with unlocked cylindrical joints and a closed structural loop is investigated. The equations of motion are derived through Maggi–Kane's method. The obtained mathematical model is free of multipliers, which makes it suitable for controller design. A trajectory planning algorithm capable of avoiding the configuration singularity of the manipulators is proposed. A slide mode controller embedded with an extended state observer (ESO) is designed for the trajectory tracking control. Numerical simulations demonstrate the effectiveness of the trajectory planning and control strategy for the reconfiguration process.

Event cameras are novel neuromorphic vision sensors with ultrahigh temporal resolution and low latency, both in the order of microseconds. Instead of image frames, event cameras generate an asynchronous event stream of per-pixel intensity changes with precise timestamps. The resulting sparse data structure impedes applying many conventional computer vision techniques to event streams, and specific algorithms should be designed to leverage the information provided by event cameras. In our work, a motion-and scene-adaptive time threshold for event data is proposed. As a parameter describing the global characteristic of the event stream, this time threshold can be used for low-level visual tasks such as event denoising and feature extraction. Based on this threshold, the normalization method for the surface of active events (SAE) is explored from a new perspective. Differently from the previous speed invariant time surface, this normalized SAE is constructed by adaptive exponential decay (AED-SAE) and can be directly applied to the event-based Harris corner detector. The proposed corner detector is evaluated on real and synthetic datasets with different resolutions. The proposed algorithm exhibits higher accuracy than congeneric algorithms and maintains high computational efficiency on datasets with different resolutions and texture levels.

A recursive decentralized control scheme is presented for the trajectory tracking issue of flexible space manipulator. In the traditional decentralized control for manipulators, each link in the manipulator is viewed as an isolated second-order subsystem. Its interconnections with other links are viewed as disturbances. As a consequence, the obtained decentralized control has a simple structure but lacks high control performance. In this paper, the manipulator is considered as a set of connected second-order subsystems. The interconnections between two adjacent links are revealed by the recursive kinematics and dynamics. In such a manner, the nominal value of the interaction forces is constructed and used for compensation to improve the tracking accuracy. Then, decentralized control is devised for each link to achieve trajectory tracking and vibration suppression of the flexible manipulator. The actuators are chosen as the joint motors and the distributed force or torque actuators on the flexible links. Under the assumption that the modal information of the flexible links is obtainable by the modal filter technique, an adaptive term in the controller is deduced for estimating the bound of the model uncertainties. Moreover, a steering logic is devised for the actuators to produce the required control. The obtained controller inherits the simple structure of the decentralized control and also achieves a higher tracking performance. The numerical examples of a free flying flexible manipulator are given to illustrate the effectiveness of the proposed control scheme.