University of Sussex
Browse
Bigge,_William_Tudor.pdf (6.33 MB)

The programmable spring: towards physical emulators of mechanical systems

Download (6.33 MB)
thesis
posted on 2023-06-07, 15:40 authored by William Tudor Bigge
The way motion is generated and controlled in robotics has traditionally been based on a philosophy of rigidity, where movements are tightly controlled and external influences are ironed out. More recent research into autonomous robots, biological actuation and human machine interaction has uncovered the value of compliant mechanisms in both aiding the production of effective, adaptive and efficient behaviour, and increasing the margins for safety in machines that operate alongside people. Various actuation methods have previously been proposed that allow robotic systems to exploit rather than avoid the influences of external perturbations, but many of these devices can be complex and costly to engineer, and are often task specific. This thesis documents the development of a general purpose modular actuator that can emulate the behaviour of various spring damping systems. It builds on some of the work done to produce reliable force controlled electronic actuators by developing a low cost implementation of an existing force actuator, and combining it with a novel high level control structure running in software on an embedded microcontroller. The actuator hardware with its embedded software results in a compact modular device capable of approximating the behaviour of various mechanical systems and actuation devices. Specifying these behaviours is achieved with an intuitive user interface and a control system based on a concept called profile groups. Profile group configurations that specify complex mechanical behaviours can be rapidly designed and the resulting configurations downloaded for a device to emulate. The novel control system and intuitive user interface developed to facilitate the rapid prototyping of mechanical behaviours are explained in detail. Two prototype devices are demonstrated emulating a number of mechanical systems and the results are compared to mechanical counterparts. Performance issues are discussed and some solutions proposed alongside general improvements to the control system. The applications beyond robotics are also explored.

History

File Version

  • Published version

Pages

226.0

Department affiliated with

  • Informatics Theses

Qualification level

  • doctoral

Qualification name

  • dphil

Language

  • eng

Institution

University of Sussex

Full text available

  • Yes

Legacy Posted Date

2011-01-25

Usage metrics

    University of Sussex (Theses)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC