【正文】 gripper aware systems research。 tendon layout prone coupling torque and motion. These factors have increased the gripper grab control of the difficulty and plexity. Linkage Transmission In addition to outside the tendon drive, another one is using the link transmission mode. Which is characterized by planar linkage transmission, the advantage of good stiffness, large output, load capacity, manufacturing easy, easy to get higher accuracy, the contact between the ponents can rely on geometric closed to achieve, better able to achieve multi species and the trajectory movement of the request, but the design is plicated. Other Transmission Gripper drive will rotate through the threaded into a linear motion, pull the drive and driven by a spring between the fingers Finger generate action, the finger part adopts lever connected to each finger movements independent of each other, with a variety of crawling configuration, and the other multifingered hand pared to drive more flexible, but the longer the closing finger, the finger structure is relatively plex and prone to failure. Artificial Muscle Drive Hydraulic drive and pneumatic drive rise in recent years as an important drive, is to simulate a human muscle drive, due to material and technical constraints, these artificial muscle technology is still far from satisfying the robot gripper reliable, fast, and accurate capture function. Shape Memory Alloy Drive In addition to hydraulic and pneumatic driven gripper outside, there is a shape memory alloy drive gripper. There are four fingers, each finger need 12 drives. Characterized by load capacity, but there fatigue and life issues, finger reaction rate is not high. owe DriveAdvances from the robot gripper, the multifingered robot hand is still the most studied, but based on multifingered dexterous hand tendons and pulleys (or hose) transmission mode causes output is small, the load capacity is poor, control plexity, restricted robots in the service of its application. Despite the use of some new technologies, such as linear motors, etc., but its fundamental drive and transmission mode has not changed, still unable to overe their weaknesses. For space robots, underactuated space robot gripper is an important research direction. By underactuated finger mechanism, mechanical limit to achieve unpowered joints and springs on the shape of the passive adaptive crawling with little driving element, grab a wide range of simple control, large output, load capacity and good features. Intelligence BiosensorSpace robot gripper sensor is internal and external environmental information for the primary means to enrich the perception of the job is to raise the level of robotics and autonomous capability necessary conditions. The study includes the selection and configuration of the sensor, the new sensor design, multisensor integration and information integration. Configured on the gripper sensors include haptic and visual sensors, in addition to feel close to sensors, distance sensors. Visual and haptic space robot gripper is the most important information. General gripper handeye system will integrate and force sensor has a wealth of information through visual information can be obtained gripper global state, and signal processing and data fusion based on the results of the various parts of the gripper determine the current status and Reliable crawl. Autonomous robots crawl the study, mostly for visionbased control, such as visual positioning and servo. Work more representative Literature. But they do not work in the case of vision can not be adjusted online pose. Information Fusion and handlingToday, in the development of intelligent robots is an important issue facing the unstructured environment in which the inaccuracy of the model and its own brings uncertainty. Resolve this uncertainty is the key to the development of intelligent sensing system. At the same time, enhance the robot in a plex environment is to improve the perception of the robot autonomy and adaptability of the main method. The processing of information is the sensing robot autonomously functioning of the most important conditions. Research and development of the robot, a key unresolved issue is the development of robotic sensing and perception systems needed to enable them to unknown plex environment in action. However, due to a variety of sensors are different in nature, their effectiveness both in time and space there are differences, which means that some traditional sensor integration and fusion method is not very suitable, for Bradley and Pradeep right TraditionalSensors made integrated approach is proposed to improve resolution and visual force concept of deposition so that the robot can be controlled in a different manner depending on the sensing module to toggle. crawl control and decisionmaking systems Stability crawlResearch multifingered robot gripper crawl purpose is to explore the mechanism of human crawl and eventually developed a body of any shape can crawl, operation and use of tools to plete a variety of tasks crawl, crawl simulate the behavior of proposed staffing manpower. So far, the coordination of multifingered robot gripper operation and control technology can not meet the requirements of the operation fine, really practical from a certain distance. Crawl stability studies largely explore the mechanism of multifingered crawl, crawl subtle analysis of multifinger movements and forces, and stability analysis is the basic work grab control. Specific work is theoretical analysis, summed able to adapt to the general conditions of the algorithm. Planning and Decision Support SystemAutonomous grasp planning is based on a variety of sensor information gripper, and the host does not need to municate, independent planning and decision making. For example, special gripper sensor extravehicular local autonomy, according to the independent judgment gripper s