- Title
- Probabilistic gradient ascent with applications to bipedal robotic locomotion
- Creator
- Budden, David; Walker, Josiah; Flannery, Madison; Mendes, Alexandre
- Relation
- Australasian Conference on Robotics and Automation (ACRA 2013). Proceedings of the Australasian Conference on Robotics and Automation (Sydney 2-4 December, 2013)
- Relation
- http://www.araa.asn.au/conferences/acra-2013/table-of-contents/
- Publisher
- Australian Robotics and Automation Association (ARAA)
- Resource Type
- conference paper
- Date
- 2013
- Description
- Bipedal robotic locomotion is an emerging field within the multi-billion dollar robotics industry, with global initiatives (such as RoboCup, FIRA and the DARPA Robotics Challenge) striving toward the development of robots able to complete complex physical tasks within a human-engineered environment. This paper details the redevelopment of an omni-directional walk engine for the DARwIn-OP, with an improved online optimisation framework developed for 13 of its internal parameters. Applying two well-known optimisation algorithms within this framework yields significant improvement in walk speed and stability. A new non-convex optimisation algorithm (Probabilistic Gradient Ascent) is derived from a reinforcement learning framework and applied to the same task, yielding an average speed improvement of 50.4% and setting a new maximum speed benchmark of 34.1 cm/s.
- Subject
- bipedal robotic locomotion; robotics industry; gradient ascent
- Identifier
- http://hdl.handle.net/1959.13/1340903
- Identifier
- uon:28607
- Identifier
- ISBN:9780980740448
- Language
- eng
- Full Text
- Reviewed
- Hits: 1350
- Visitors: 1747
- Downloads: 199
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT02 | Publisher version (open access) | 302 KB | Adobe Acrobat PDF | View Details Download |