Brain-inspired Bayesian perception for biomimetic robot touch

Lepora, Nathan; Sullivan, Charlie; Mitchinson, Ben; Pearson, Martin; Gurney, K.; Prescott, Tony J.

Abstract

Studies of decision making in animals suggest
a neural mechanism of evidence accumulation for competing
percepts according to Bayesian sequential analysis. This model
of perception is embodied in a biomimetic tactile sensing
robot based on the rodent whisker system. We implement
simultaneous perception of object shape and location using two
psychological test paradigms: first, a free-response paradigm
in which the agent decides when to respond, implemented
with Bayesian sequential analysis; and second an interrogative
paradigm in which the agent responds after a fixed interval,
implemented with maximum likelihood estimation. A benefit
of free-response Bayesian perception is that it allows tuning
of reaction speed against accuracy. In addition, we find that
large gains in decision performance are achieved with unforced
responses that allow null decisions on ambiguous data. Thus,
free-response Bayesian perception offers benefits for artificial
systems that make them more animal-like in behavior.

Citation

Lepora, N., Sullivan, C., Mitchinson, B., Pearson, M., Gurney, K., & Prescott, T. J. (2012, May). Brain-inspired Bayesian perception for biomimetic robot touch. Paper presented at IEEE International Conference on Robotics and Automation, Saint Paul, Minnesota, USA

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: IEEE International Conference on Robotics and Automation
• Conference Location: Saint Paul, Minnesota, USA
• Start Date: May 14, 2012
• End Date: May 18, 2012
• Publication Date: Jan 1, 2012
• Deposit Date: Jan 21, 2013
• Publicly Available Date: Feb 11, 2016
• Peer Reviewed: Peer Reviewed
• Keywords: Bayesian perception, biomimetic robot touch
• Public URL: https://uwe-repository.worktribe.com/output/954397
• Additional Information: Title of Conference or Conference Proceedings : IEEE International Conference on Robotics and Automation (ICRA)