The cocktail party problem is a well-known challenge in auditory perception, yet it is still unclear how bats are able to overcome it in complex acoustic environments. This thesis addresses this question from a robotic perspective, translating the problem into an echolocating robotic context to investigate it experimentally.
The work presented in this thesis focused on scaling the RO-BAT platform, both in hardware and software, from a single prototype of an echolocating robot capable of obstacle detection to a group of robots capable of multi-target tracking through Bayesian filtering techniques.
The resulting framework enables the experimental investigation of acoustic jamming and the cocktail party problem in echolocation, providing the basis for studying these effects in a controlled robotic environment.
About the speaker
Riccardo Corà earned a Bachelor’s degree in Computer Engineering from Politecnico di Milano in 2023 and later pursued a Master’s degree in Acoustic and Music Engineering at Politecnico di Milano. He completed his Master’s thesis at the Center for the Advanced Study of Collective Behaviour department of the University of Konstanz within the RO-BAT project, scaling the platform from a single bio-inspired echolocating robot capable of obstacle avoidance to a multi-robot setup relying on Bayesian filtering for tracking multiple echoes.
