The Development of System Identification Approaches for Complex Haptic Devices and Modelling Virtual Effects Using Fuzzy Logic

Systems Design;System Identification;Haptic;Virtual Wall;Fuzzy Logic

Tam, Sze-Man Samantha

University of Waterloo

关键词: Systems Design;    System Identification;    Haptic;    Virtual Wall;    Fuzzy Logic;   

Others  :  https://uwspace.uwaterloo.ca/bitstream/10012/924/1/s2tam2005.pdf

瑞士|英语

来源: UWSPACE Waterloo Institutional Repository

【 摘 要 】

Haptic applications often employ devices with many degrees offreedom in order to allow the user to have natural movement duringhuman-machine interaction.From the development point of view, thecomplexity in mechanical dynamics imposes a lot of challenges inmodelling the behaviour of the device. Traditional systemidentification methods for nonlinear systems are oftencomputationally expensive. Moreover, current research on usingneural network approaches disconnect the physical device dynamicswith the identification process.This thesis proposes a differentapproach to system identification of complex haptic devices whenanalytical models are formulated. It organizes the unknowns to beidentified based on the governing dynamic equations of the deviceand reduces the cost of computation. All the experimental work isdone with the Freedom 6S, a haptic device with input and feedback inpositions and velocities for all 6 degrees of freedom .

Once a symbolic model is developed, a subset of the overall dynamicequations describing selected joint(s) of the haptic robot can beobtained. The advantage of being able to describe the selectedjoint(s) is that when other non-selected joints are physically fixedor locked up, it mathematically simplifies the subset dynamicequation. Hence, a reduced set of unknowns (e.g. mass, centroidlocation, inertia, friction, etc) resulting from the simplifiedsubset equation describes the dynamic of the selected joint(s) at agiven mechanical orientation of the robot. By studying the subsetequations describing the joints, a locking sequence of joints can bedetermined to minimize the number of unknowns to be determined at atime. All the unknowns of the system can be systematicallydetermined by locking selected joint(s) of the device following thislocking sequence. Two system identification methods are proposed:Method of Isolated Joint and Method of Coupling Joints. Simulationresults confirm that the latter approach is able to successfullyidentify the system unknowns of Freedom 6S. Both open-loopexperimental tests and close-loop verification comparison betweenthe measured and simulated results are presented.

Once the haptic device is modelled, fuzzy logic is used to addresschattering phenomenon common to strong virtual effects. In thiswork, a virtual wall is used to demonstrate this approach.Thefuzzy controller design is discussed and experimental comparisonbetween the performance of using a proportional-derivative gaincontroller and the designed fuzzy controller is presented. Thefuzzy controller is able to outperform the traditional controller,eliminating the need for hardware upgrades for improved hapticperformance. Summary of results and conclusions are included alongwith suggested future work to be done.