Legged robots are well-suited for navigating unstructured terrain due to their ability to perform agile dynamic motions. Building on the efforts of Cal Poly's Legged Robotics Group, this project focuses on achieving stable forward hopping with a single legged robot.
A comprehensive model of leg hopping is formulated, with leg dynamics during flight and stance phases derived using Euler-Lagrange equations. During the stance phase, a spring-mass system models the leg-ground interaction. The touchdown and lift-off conditions are formulated mathematically, and state mappings between phases are determined via momentum conservation. A two-phase hybrid controller is designed, incorporating trajectory planning during flight and ground-reaction force and impedance control during stance. Numerical simulations validate the design, followed by experimental testing on a Speedgoat real-time machine within a Simulink environment, showing good correlation with simulations. The impact of a control parameters on average velocity are experimentally analyzed.
The results demonstrate the controller's effectiveness in achieving stable hops and provide a method for tuning hop speeds. This research also paves the way for future projects by thoroughly documenting the leg's physical parameters and addressing the challenge of implementing reliable impact and lift-off detection during experimental testing.
| Identifer | oai:union.ndltd.org:CALPOLY/oai:digitalcommons.calpoly.edu:theses-4568 |
| Date | 01 August 2024 |
| Creators | Bennett, John Anthony |
| Publisher | DigitalCommons@CalPoly |
| Source Sets | California Polytechnic State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Master's Theses |
Page generated in 0.0014 seconds