Space Robot Fault Detection Diagnosis And System Reconstruction

As the complexity and investment of spaceflight system is increasing, the life-span and reliability is also required to be very high. The spaceflight system should posses the ability of fault tolerant, detect and diagnosis where is the fault, and reconstruct itself automatically. Space robot is the important part of large scale spaceflight system, such as space laboratory and space shuttle, it can capture and release effective load, and is very efficient to the connection of shuttles. As a kind of large scale space mechanism, the life-span of space robot should be long, so it is required to work high reliably, its joints should cooperate with each other, and it can detect exactly where is the fault.Every aspect of space robot fault detection and system reconstruction is researched systemically, including system design, fault detection, fault tolerant design and inverse kinematics trajectory control. The results are as follows.First, the elementary and brief principle of space manipulator structure design is set up. This is the basis of following research.Second, the fault detection methods of analog circuit, digital circuit and control system are introduced, especially the technology of sneak circuit analysis, and two new sneak circuit analysis methods are introduced, they are key node based sneak circuit analysis and qualitative reasoning based sneak circuit analysis. The two methods are effective to shorten analysis hours, improve efficiency, and reduce costs.Third, the theory of general purpose fault tolerant manipulator design and task specified fault tolerant manipulator design is introduced. Optimize design methods of fault tolerant manipulator are found. A kind of fault tolerant manipulator possessing least degrees of freedom is work out.Fourth, the trajectory control methods of inverse kinematics joints space of all kinds of redundancy manipulators are analyzed. A kind of null space optimization based multi-condition restricted inverse kinematics methods of redundancy manipulator is introduced. Another kind of inverse kinematics methods of redundancy manipulator is introduced, if s restrict conditions are expressed in general vectors, if s tasks are organized according to their importance, it' s kinematics singularity robust, it' s arithmetic singularity robust, if s fuzzy theory based multi-task coordinated. At the same time, the feasibility of manipulator reconstruction method is also validated. All these lay a foundation for the kinematics control of space manipulator.