A comparison of analytical and empirical controllers for the SLIP hopper

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The Spring-Loaded Inverted Pendulum (SLIP) model poses a challenging control problem, important for the development of legged robots, due to the difficultly in solving the stance phase of the dynamics. Multiple attempts have been made to approximate these dynamics to allow for an analytical control method; here four of these methods have been compared for controlling agile hopping, where there are large changes in forward velocity across a single stance. In addition, a new, empirical, approach has been demonstrated. In this, a simple control law is formulated, based on some simple approximations, which allows the parameters to be selected empirically through simulation. This has led to a controller able to offer similar performance to the best analytical approximation but with a much simpler form. This empirical controller may present new opportunities for controlling more complex dynamics and the development of a self-tuning method in future work.

Original languageEnglish
Title of host publicationTowards Autonomous Robotic Systems - 18th Annual Conference, TAROS 2017, Proceedings
Place of PublicationGermany
PublisherSpringer Verlag
Pages79-85
Number of pages7
ISBN (Print)9783319641065
DOIs
Publication statusE-pub ahead of print - 20 Jul 2017
Event18th Annual Conference on Towards Autonomous Robotic Systems, TAROS 2017 - Guildford, UK United Kingdom
Duration: 19 Jul 201721 Jul 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10454

Conference

Conference18th Annual Conference on Towards Autonomous Robotic Systems, TAROS 2017
CountryUK United Kingdom
CityGuildford
Period19/07/1721/07/17

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'A comparison of analytical and empirical controllers for the SLIP hopper'. Together they form a unique fingerprint.

Cite this