Robotics-1: Event-based vision, navigation, and SLAM

Yulia Sandamirskaya

In this session, we will discuss the role of neuromorphic computing in robot navigation. The discussion will start with 5 short presentations (statements) from prominent researchers, who already have successfully applied neuromorphic computing in their work – either in form of event-based vision algorithms running on a conventional computer or SNNs running on Loihi. We will then have a panel discussion, aiming to narrow down answers to the following questions:

• Which subtasks in robot (vehicle, drone) navigation are particularly attractive for neuromorphic solutions and which are not? What would be the criteria for success of a solution? Which bottleneck could neuromorphic realization solve: scalability? latency? throughput?

• What are the challenges to make Neuromorphic approaches enter the main stream in robot navigation: Algorithms? Interfaces? Performance? Ecosystem?

Other questions will be collected prior to the workshop.

• Prof. Davide Scaramuzza (Univ. Zurich)

  • Event-based vision for high-speed control of drones

• Dr. Tobias Fischer (QUT)

  • Use of event cameras for the visual place recognition task

• Prof. Konstantinos Michmizos (Rutgers Univ.)

  • Robotic Navigation: Scaling Brain-Inspired Computing by pursuing its biological plausibility

• Prof. Guido de Croon (TU Delft)

  • Neuromorphic sensing and computing for tiny drones: Opportunities and challenges.

• Dr. Torben Schoepe (Univ. Groningen)

  • Insect-inspired robot navigation

• Max Ruffo (Terabee)

  • Lean sensing for mobile robots

Link to a teaser for the guided discussion material coming soon. Please, add your questions and topic that you would like to discuss in the comments below.

Not yet available

Not yet available