Contact: Dümbgen, Frederike
Synopsis: Audio-based source tracking and obstacle avoidance on the e-puck robot
Level: BS/MS semester project
Description: When compared to visual and inertial measurements, sound is highly underrepresented among the modalities commonly used for navigation, localization and mapping in robotics. This is surprising as animals such as bats and whales have evolved to use this modality very efficiently for a variety of tasks such as prey localization, tracking and obstacle avoidance.
In this semester project, you will port multiple audio-based localization algorithms of our lab, originally designed for drones, to the e-puck2 robot. You will test the performance of these algorithms in different scenarios and compare them with classical and state-of-the-art methods. In the end, you exploit these algorithms to implement a little demo of e-puck robots doing source tracking, wall following and/or swarming.
Deliverables: A technical report and a clean code base.
Prerequisites: Good programming skills, experience with audio and the e-puck robot are a plus.
Type of Work: 20% theory/research, 50% coding, 30% experimental validation.