Font Size: a A A

Research On Modular Force/Position Control For Environmental Constrained Reconfigurable Manipulator

Posted on:2016-06-09Degree:MasterType:Thesis
Country:ChinaCandidate:Y SongFull Text:PDF
GTID:2308330467498905Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Reconfigurable manipulators are able to utilize some independent joint or connectingrod modules of various sizes and properties. By adding, removing or exchanging modules toassemble the required configuration, this reconfiguration includes not only the reconstructionof mechanical structure, but also that of the control system, so that the reconfigurablemanipulators possess good structural flexibility. Compared with traditional manipulators,reconfigurable manipulators are such flexible, low-cost, stable that they can complete morecomplicated tasks and adapt more quickly to different working environment. Because ofthese very advantages, the reconfigurable manipulators have broad prospect of application inmanufacturing, national defense, aerospace, search and rescue, medical services and businessservices. Hence, researches on reconfigurable manipulators are of great importance both intheory and application.When reconfigurable manipulators are conducting some tasks like precise assembly,crawling, handling etc. under specific circumstances, the manipulator will inevitably comeinto contact with the environment or manipulating objects, which can cause constrains onmanipilators. At this time, it is out of requirement by simply controlling the position ofmanipulator, the contact force on the end also need to be controlled in order to avoidcollision-caused damages on the end of manipulator or manipulated object or the contactsurface. Therefore, to meet the requirements of it is of urgent necessity to address theproblem of controlling force/position on the reconfigurable robotic arm under constrainedcircumstances. Considering the modular property and structural uncertainty ofreconfigurable manipulator, traditional control methods are bound to constrain the flexibilityof the reconfigurable manipulator. Because of the high efficiency in solving the problem ofcontrolling high dimensional, and strong coupling system, modular control structure mayquite suitable for the idea of modular reconfigurable manipulators. Besides, it is a premise toresolve the reflecting relationship between the force on the end-effector and the joint force/torque, if we want to solve the problem of modular force/position control on themanipulators. From the data collected at present, researches on the problem of modularforce/position control on reconfigurable manipulators are hardly reported. In this paper,focusing on the reconfigurable manipulators at constrained circumstances, we researched onthe methods of dynamic modeling of the manipulators, the hybrid control method ofself-adaptive fuzzy force/position under unknown disturbances, and deduced the reflectingrelationship between the force on the end-effector and the force/torqueon the joints, found amodular force/position control methord relying only on local joint information.The main contents are as follows:1. Explained the background and significance of our research; made a general narrativeon the researching state, hot problems, and methods of reconfigurable manipulators;2. Taking into account the force/torque acting on each joint and connecting rod, and thatbetween the joints, we built a dynamic model by connecting movement and the force/torqueof the manipulator, through an effective set of both positive and negative direction iterativeequation based on Newton-Euler iterative algorithm; Deducing the reflecting relationshipbetween the force on the end-effector and the force/torqueon the joints, obtained thedynamical model of environmental constrained reconfigurable manipulator.3. Based on the reduced dynamics, we investigated the problem of self-adaptive fuzzyforce/position control for environmental constrained reconfigurable manipulators. The fuzzylogic systems are designed to approximate the overall dynamics, and unknown disturbance isovercome by adaptive algorithm. The effect of the approximation error is removed byemploying an adaptive sliding mode compensation term. Finally, we evaluated the stabilityof designed controller through Lyapunov stability theorem, and verified its effectiveness andversatility via simulation.4. On the environmental constrained reconfigurable manipulator system, we invented amodular force/position control methord which relies only on the local joint information. Onthe condition of wrist force sensor being installed at the end of the manipulator, wedecomposed the original dynamic model into the form of a set of subsystem and couplingassociated items, reflect the contact force between the end and the environment into each subsystem through Jacobian matrix, and control the position and force of end-effector bycontrolling the position and torque of each subsystem. An approximation to the nonlinearterm and interconnection term was achieved by using neural network. The effect of theapproximation error was removed by employing an adaptive sliding mode compensationterm. Finally, the system stability was assured by the Lyapunov stability theorem, theeffectiveness and versatility of the method was ascertained by simulation.5. Made a conclusion about the whole article, and gave a prospect on research work inthe future.
Keywords/Search Tags:Reconfigurable manipulator, Environmental constrained, Modular control, Force/position control, Adaptive fuzzy, Neural network
PDF Full Text Request
Related items