Force control and collision avoidance strategies for kinematically redundant manipulators
The problem of position control of non-redundant manipulators was addressed during the initial stages of development of robotics in the 70's. In the 80's, extension of robotic applications to new non-conventional areas, such as space, underwater, hazardous environments, and microrobotics,...
| Main Author: | |
|---|---|
| Format: | Others |
| Published: |
1997
|
| Online Access: | http://spectrum.library.concordia.ca/607/1/NQ40320.pdf Shadpey, Farshid <http://spectrum.library.concordia.ca/view/creators/Shadpey=3AFarshid=3A=3A.html> (1997) Force control and collision avoidance strategies for kinematically redundant manipulators. PhD thesis, Concordia University. |
| Summary: | The problem of position control of non-redundant manipulators was addressed during the initial stages of development of robotics in the 70's. In the 80's, extension of robotic applications to new non-conventional areas, such as space, underwater, hazardous environments, and microrobotics, brought new challenges for robotic researchers. Position control strategies failed in performing tasks that needed interaction with a robot's environment. On the other hand, non-redundant manipulators were unable to perform tasks that required dexterity comparable to that provided by the human arm. Also, imprecise dynamic modeling put severe restrictions on performance of control algorithms which were based on exact knowledge of dynamic parameters. These issues have therefore attracted a lot of attention in following three areas: force and compliant motion control, redundancy resolution, and adaptive control strategies. These areas have been addressed separately. However, there exists no unique frame work for an adaptive compliant motion control scheme for redundant manipulators which enjoys all the desirable characteristics of the methods that have been proposed for each individual area, e.g., the existing compliant motion control schemes are either not applicable to redundant manipulators or cannot take full advantage of the redundant degrees of freedom. In this thesis, the existing schemes in each of these three areas are reviewed. Based on the results of this review, a new redundancy resolution scheme at the acceleration level is proposed. The feasibility of this scheme is studied using simulations on a 3-DOF planar arm. This scheme is then extended to the 3-D workspace of a 7-DOF redundant manipulator. The performance of the extended scheme with respect to static and moving object collision avoidance and also joint limit avoidance is studied using both simulations and hardware experiments on REDIESTRO (a REdundant, Dextrous, Isotropically Enhanced, Seven Turning-pair RObot constructed in the Center for Intelligent Machines at McGill University). Based on this redundancy resolution scheme, an Augmented Hybrid Impedance Control (AHIC) scheme is proposed. The AHIC scheme provides a unified frame work for combining compliant motion control, redundancy resolution, and adaptive control in a single methodology. The feasibility of the proposed AHIC scheme is studied by computer simulations and experiments on REDIESTRO. |
|---|