Bio-inspired SLAM

Simultaneous localization and Mapping a.k.a. SLAM has been a well researched topic in the robotics community for over two decades. However, with the recent advances in the Neuroscience yielding deeper insights into the neural correlates of the mammalian brain, there is renewed interest in further exploration of nature inspired SLAM approaches.

Rats, bats, and other mammals like humans are able to navigate even in unknown environments and with mammals especially like rats that posses a lightweight computational capability, there is scope for exploring alternatives to Deep Learning solutions to SLAM.

Thus, in this project, we investigate neural networks that mimic the activation of rat brains for performing SLAM using biomimetic sensors like coarse camera and sparse tactile sensors.

People involved

Project updates

18. September 2019: Presenting our Work “ViTa-SLAM: A Bio-inspired Visuo-Tactile SLAM for Navigation while Interacting with Aliased Environments” at the IEEE conference on Cyborg and Bionic Systems (CBS). In this conference paper we established our work on ViTa-SLAM, a visuo-tactile extension of RatSLAM that allows a robot to distinguish visually ambiguous scenes by interacting with the environment through touch.

20. May 2019: Presenting a Workshop paper titled “ViTa-SLAM: Biologically-Inspired Visuo-Tactile SLAM” at the ICRA 2019 ViTac Workshop. In this workshop paper we first published ViTa-SLAM, our framework for visuo-tactile SLAM.