Search

Active-Passive Hybrid Cable-Driven Robots (APHCDRs) are a novel type of cable-driven robots with the design of active-passive-linkage segments. Compared with tradition ones, they are more flexible and have higher maneuverability in confined environment, thus, are widely researched in recent years. However, the workspace and stiffness performance of APHCDRs which are two important properties were rarely studied. In this paper, the generalized kinematic model of APHCDRs is established first. Then, the resolved workspace and stiffness model are proposed. Workspace radius and stiffness are analyzed by considering link length and linkage ratio respectively. Finally, particle swarm optimization (PSO) is used to search for the optimal configurations of APHCDRs which reach maximum workspace area, working radius and maximum stiffness respectively. The analysis results on the workspace and stiffness are useful for optimizing of APHCDRs’ structural design.

A cable-driven hyper-redundant manipulator receives great attention due to its slender body and flexible movement. The configuration of the driving cable is the key to manipulator design. At present, cables are arranged at equal angles on the disk, but it lacks theoretical support. The analysis about the influence of cable positions on cable tensions is relatively deficient. In this paper, the force on links is studied. The influence of cable positions on driving force is analyzed. The relationship between cable position, driving force, external force and joint angles are studied. Numerical simulation is carried out to verify the theoretical analysis. The results show the maximum of driving forces is the smallest when cables are arranged at equal angles and decreases as the cable position becomes uniform. The results explain the rationality of the equal configuration of the cables in the cable-driven hyper-redundant manipulator. It’s beneficial to guide the design of cable-driven manipulator.

A hybrid active and passive cable-driven segmented hyper-redundant manipulator is very flexible and dexterous to conduct tasks in highly cluttered environment. However, computation load of inverse kinematics and trajectory planning are also very large. In the paper, a kinematic equivalence method is proposed for the hybrid active and passive cable-driven segmented hyper-redundant manipulator to overcome the above challenge when the position and direction of end-effector are considered. The kinematic equivalence method is an effective way to solve the inverse kinematics and trajectory planning by simplifying and rearranging joints of each segment. The mechanism and joint layout of the manipulator are first analyzed. Then, the kinematics model is established by both traditional DH method and kinematic equivalence method. The calculated amount is decreased by reducing the number of rotation axis that needs to be processed in each segment. Furthermore, the desired trajectory is generated for the end effector of the arm to approach the target point. Finally, the proposed method is applied to a practical prototype, which has five segments and each segment consists of six subsegments. Simulation results verified the proposed method.

Pose (position and attitude) measurement of a space non-cooperative target is very important for on-orbital servicing tasks, including malfunctioning target repairing, space debris removal, and so on. However, such targets are generally non-cooperative, i.e., no markers are mounted on it and there is no prior knowledge. Therefore, the identification and measurement of a non-cooperative target is very challenging. In this paper, we propose an efficient method to recognize the natural objects with circular or near-circular shapes on the target, reducing the computation load and improving accuracy. First, the geometry properties of a practical non-cooperative target are analyzed. A stereo vision system is correspondingly designed to measure the relative pose of the target. Second, the error sources and time-consuming factors of the traditional method are analyzed. Then, a solution concept is proposed. Third, the efficient method is detailed to solve the geometry equation and determine the pose information, reducing the calculation complexity and increasing the accuracy. The image pre-processing and target detecting algorithms are realized on FPGAs, further accelerating the calculation speed. Finally, we develop an experiment system and verify the proposed method through practical experiments. The experiment system is composed of satellite mockup, binocular camera, and high precision laser tracker. The experiment results show that the proposed method has high accuracy and efficiency.

The pose (position and attitude) measurement of a space un-controlled target is very important for on-orbit service. However, there are no cooperative markers on these targets, so it is difficult to identify and measure of such targets. In this paper, an efficient VSVM (virtual stereo-vision measurement) method is proposed, which can recognize the natural objects with the circular and triangle features on the target, and largely reducing the computation load. Firstly, based on 3D reconstruction, the 3D point cloud feature of non-cooperative target is reconstructed. According to the 3D point cloud information, the relative pose of the satellite nozzle (circular center) and the triangular bracket (the center of the inscribed circle in the triangle) can be obtained accordingly. Secondly, a stereo-vision measurement method of circular features and triangular features is derived. Thirdly, the efficient VSVM method is used to solve the pose coupling equation and obtain the pose information, reducing the number of sensors and the calculation complexity; it also reduces the cost of space measurement. Finally, an experimental system is built and the validity of the algorithm is verified by the practical experiments. The experiment system is composed of two cameras, a FARO laser tracker, and satellite mockup. The experiment results demonstrate the effectiveness of the proposed method.

A malfunctioned satellite or other space debris bringing great challenge to capture and remove it. In this paper, we propose an efficient trajectory planning method for rapid capturing of a space tumbling target. Firstly, the kinematics model of a dual-arm space robot is deduced in details. Secondly, the motion characteristics of a tumbling target are then analyzed. Thirdly, an efficient trajectory planning method for rapid capturing of the space tumbling target is presented. It improves the speed of the capturing process. Finally, the simulation environment is created and the proposed method is verified. The simulation results show that the proposed trajectory planning method of the dual-arm space robot is effectively for tumbling target capturing. The proposed method is very important for the future applications.

Due to uncontrolled, a malfunctioned satellite or other space debris is generally tumbling, bringing great challenge to capture and remove it. In this paper, we propose an autonomous capturing strategy based on impedance control to detumbling the target. Using this strategy, the end-effectors will approach the capture point on the docking ring of target and then stabilize the tumbling target. An un-controlled space object and a space robotic system are first modeled. Dynamics model of tumbling target is established with or without external forces and torques acting on the target. And the motion characteristics of target is theoretically analyzed when there are no external forces and torques impacting on the target. Then, the capture and detumbling strategy is presented base on impedance control. The strategy has two phases. The first is approaching phase that the trajectory is planned for the arm to approach to capture point. The other one is detumbling phase that the tumbling target is stabilized. Finally, we develop an Adams-Simulink co-simulation system. Based on it, the proposed method is verified by space robotic system. Simulation results show that the end-effector will approach the capture point on the docking ring of target rapidly and then stabilize the tumbling target gradually.

Imaging based target tracking by a robotics is an interesting and popular topic. While the high computational power requirement of imaging processing algorithm is still a big limitation, some flying robot like micro aerial vehicles also has a payload limitation, so the high weight computer and GPU is no longer a good choice. This paper present a high speed ball target tracking system based on FPGA. With a CMOS sensor, the original frame can be transfer to the FPGA port. After the video streams go through the parallel image processing core, the color and corner features can be extracted. By comparing with the predefined target's pattern, the target can be detected. AXI bus can transfer the detected target's feature to the ARM core in the ZYNQ device. The PID controller embedded in ARM core can be used to control the Pan-tilt header to move the target to the center of frame.

With regard to Automated Guided Vehicles (AGVs) and autonomous navigation of Micro Aerial Vehicles (MAVs) in complicated environments, it is vital to detect not only in the direction of ongoing but also the entire environment. It is very significant for obtaining a broader view of a scene than the current single view which has been used in a wide range of applications such as satellite imaging, medical imaging and so on. As the conventional omnidirectional camera based on PC is cumbersome and power consuming, then a novel light-weight structure comprising of four cameras to detect targets in all directions based on FPGA is presented in our design. We will discuss the implementation of image mosaic algorithm which is comprised of image registration and image fusion and some other image preprocessing algorithms for better effect such as median filter algorithm, color filter algorithm, image enhancement algorithm, etc., on the Xilinx Zynq-7020 FPGA device using Vivado Design Suite and Software Development Kit which is embedded on ZedBoard developed board. The system is able to handle more than 60 frames per second (fps) freely and still in a low power consuming.

A parallel hardware architecture for real-time image feature detection based on the SIFT (Scale Invariant Feature Transform) algorithm has been presented. The proposed parallel hardware architecture is completely stand-alone; it reads the input data directly from VITA2000 and provides the results via a Field-Programmable Gate Array (FPGA); image key points are extracted in combination with SIFT IP Core. Our proposed parallel hardware system could be able to detect feature points up to 30 frames per second with the resolution 1920∗1080; the proposed method has the similar accuracy to PC implementation. Experimental results shown that the hardware architecture for SIFT algorithm realizes fast feature extraction, the shortcomings of massive calculation and low speed in the process of extracting image features have been efficiently improved, which meets the real-time requirements in feature matching system. The achieved system performance is at least two orders of magnitude better than a PC-based solution; this algorithm can be applied to feature matching in the field of Robotics.

Recently, FPGA becomes more important and popular in embedded systems due to its parallelizable inner structures. In this paper, a pipelined FPGA architecture will be present which will be used on a mobile robot. It implements algorithms about target recognition, tracking, positioning based on monocular vision, etc. The hardware platform for the design is the development board by Xilinx. The system uses monocular camera to measure the orientation of target in real time. It acquires the video stream from CMOS camera. The system realizes target tracking by extracting the color of the target as the feature, filtering, detecting feature of the target and then displaying on the screen. The system realizes target positioning according to the centroid of the tracked target, then computing the orientation which can be used for the intelligent robot vision.

Compared with a single-arm space robot, a dual-arm space robot has much more dexterity and flexibility, and is capable to complete more complex tasks. The on-orbit capturing and manipulation of a target is the key of the on-orbit servicing mission. After the target is captured by the two arms, a compounded system is formed. The two arms connect the robotic base to the target, and a closed kinematics chain is formed. In this paper, we design a dual-arm space robotic system to serving the target spacecraft, investigate the topology, and derived the corresponding dynamic model. Based the dynamic equations, we developed a closed-loop simulation system. At last, we analyze the key problems for the closed-loop system for manipulating the target, including the joint locking torques, disturbance motion resulting from the centroid position changing, and control algorithm. The derived model and analysis results are very important for the future applications.

Redundant space manipulators, including Space Station Remote Manipulator System (SSRMS), Special Purpose Dexterous Manipulator (SPDM) and European Robotic Arm (ERA), have been playing important roles in the construction and maintenance of International Space Station (ISS). They have similar joint layout and are referred to SSRMS-type manipulators. This paper presents an effective inverse kinematics resolution method for this type of manipulators for single joint locked failure. Firstly, the configuration characteristics of the SSRMS-type redundant manipulators are analyzed. Secondly, the D-H frames and corresponding D-H parameters of the new 6-DOF manipulator formed by locking a joint in an arbitrary position are re-constructed. Thirdly, the inverse kinematics problems are classified into two cases: completely analytical inverse kinematics for locking joint 1, 2, 6 or 7, and hybrid inverse kinematics for locking joint 3, 4 or 5. The corresponding approaches are presented to solve them. Finally, some practical examples for typical single joint locking are analyzed. The results verify the proposed methods.

A grinding robot is composed of six joints, within which the second joint is a prismatic joint and the others are rotational joints. According to its configuration characteristic, the 6DOF of the inverse kinematics problem was decomposed into two 3DOF sub-problems: the position inverse kinematics and the attitude inverse kinematics. Then the two sub-problems were respectively solved analytically, and four sets of solutions of the joint variables were obtained. By adopting the configuration flag, the appropriate solution corresponding to the flag can be determined. Furthermore, the circular trajectory in Cartesian space of the grinding robot was planned to perform the grinding tasks. By taking advantage of the analytical inverse kinematics, the planned Cartesian trajectory is transformed into joint-space. Finally, a simulation system was developed by using Matlab/Simulink software. Based on the system, the analytical inverse kinematics solution and trajectory planning method was studied by numerical simulation. The simulation results showed the effectiveness of the proposed method.