Adaptive jacobian control of robot

• Main Author: Tan, Jesper Jian Xiong.
• Other Authors: Cheah Chien Chern
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
• Online Access: http://hdl.handle.net/10356/40406
• Institution: Nanyang Technological University

Adaptive depth and length controllers were implemented and tested with the use of a Selective Compliance Assembly Robot Arm (SCARA). The controllers are a set of control loop feedback mechanisms that make use of the vision feedback from a camera to track an infra-red LED that is attached to the end of the SCARA robot arm. The controllers were tested with setpoint tasks to determine the robustness of the controller. Uncertainty in depth of camera vision and length of robot’s arm were introduced to determine if the controller can adapt to the uncertainties. In the initial stage of the project, a Proportional-Derivative controller was implemented and tested to determine the robustness of the system as a reference for performance for later stages. From the experiments, it was observed that when the spring force gain, Kp, was of sufficient magnitude to move the end effector accurately to a desired position, a ratio of Kp/Kv = 4 was needed to make sure that the end effector stopped oscillating at steady state, where Kv is the damping force gain. In the second stage of the project, the adaptive depth and length controllers were implemented and tested separately to determine the effects of the control gains. The initial estimated parameters and update law’s gains were varied, and it was observed that they have a significant impact on the stability of the controller. When the initial estimated parameters were set to a value approximate to the real physical parameters of the system, the controllers exhibited higher degree of stability. The final stage of the project involved implementing both adaptive depth and length controllers together. The controller was tested by introducing uncertainty in camera’s depth of vision. The camera was tilted 10° to 20° in both x and y direction to introduce uncertainty in depth. It was observed in the experiments that the controller was able to perform point to point tracking with minimal errors even when the uncertain conditions were given. Hence it showed that the full adaptive controller can perform setpoint tasks with high degree of stability even when there were uncertainties in its physical parameters