Home
Lectures
Reinforcement Leaning Adaptive Control of a Flexible Manipulator

Reinforcement Leaning Adaptive Control of a Flexible Manipulator

Speaker: Bidyadhar Subudhi – Ponda Goa, India
Topic(s): Architecture, Embedded Systems and Electronics, Robotics

Abstract

Manipulators with thin and light weight arms or links are called as Flexible-Link Manipulators (FLMs). These manipulators offer several advantages over their rigid-link counter parts including achieving high speed operation, lower energy consumption, and increase in payload carrying capacity and large workspace. However, designing a feedback control system for a flexible-link manipulator is challenging because of its non-minimum phase behaviour, under-actuation and non-collocation. Further, the control difficulties are more complex when such manipulators are subjected to handle unknown payloads. In order to represent the dynamics of such FLMs accurately, it is necessary to consider an infinite number of flexible modes in its distributed parameter model. However, to facilitate controller realizability higher order modes are truncated. Model uncertainty due to truncation of flexible modes in its dynamics leads to inaccurate tip-tracking performance and system instability. Due to change in payload, the flexible modes are excited giving rise to uncertainties in the dynamics of the FLM. To achieve effective tip trajectory tracking whilst quickly suppressing tip deflections when the FLM carries varying payloads adaptive control is necessary instead of fixed gain controller to cope up with the changing dynamics of the manipulator. This lecture will cover design of advanced adaptive control strategies such as reinforcement learning based adaptive controller for a two-link flexible manipulator with a nonlinear autoregressive moving average exogenous (NARMAX) model identified model of FLM to control the tip trajectory tracking and its deflections while handling unknown payloads. The reinforcement learning based adaptive control (RLAC) has an advantage that it attains optimal control adaptively in on-line. The performance of the RLAC is compared with that of two other adaptive controllers e.g. Direct Adaptive Controller and Fuzzy Model Reference Adaptive Controller.

About this Lecture

Number of Slides: 45
Duration: 60 minutes
Languages Available: English
Last Updated:

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.