Qian, Wang; (1992) Study on the dynamics and force control of a robotic manipulator. Doctoral thesis (Ph.D), UCL (University College London).

Text Study_on_the_dynamics_and_forc.pdf Download (7MB)

Abstract

Self-tuning adaptive control for robotic manipulators is the main theme of this thesis and is used for dynamic control and force control of a robotic manipulator both in theoretical simulation and in experimental work. A simplified dynamics model has been developed for a PUMA type robotic manipulator. Heavy symbolic calculation has been carried out to make full use of the special PUMA geometry so as to further reduce the mathematical burden in controlling the arm dynamics. Extensive simulation has been carried out using the digital computer. A PASCAL program package with graphics display has been produced for robotic assembly (peg-into-hole) on a VAX workstation. Various dynamic control simulation programs have been written on an IBM PC using MODULA-2. A new self-tuning PID controller, whose gains have an explicit relation with process parameters, has been worked out. A new simulation scheme, which can make direct use of the Newton-Euler equations, has been developed for the robot control. The self-tuning PID controller is used for the outer-loop force control of the PUMA560 industrial robotic manipulator. A three dimensional compliant device was designed to go between the robotic end-effector and the work environment. A PUMA560 supervisory control program package, incorporating real-time compliant motion control, written in MODULA-2 was developed on an IBM PC, with menu and multi-process support. Experiment has shown that adaptive compliant motion control can largely improve contact quality and tracking ability for robotic inspection in an unknown environment. The author has also succeeded in putting a peg into a hole with a maximum clearance 0.02 mm using the PUMA560.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item