Challenges and Opportunities of Event-based Perception in Field Robotics and Automation

Context

Neuromorphic technology in robotics, such as event cameras and neuromorphic processors, have advanced significantly over the last three decades. Despite breakthroughs in hardware and algorithms, the adoption of neuromorphic technology in real-world robotic systems remains limited. Key challenges include real-time processing, managing high data rates, learning representations, and integration with control and planning frameworks. This workshop aims to bridge the gap between the neuromorphic and robotics communities, encouraging cross-disciplinary dialogue and fostering adoption of event-based control and sensing in field robotics and automation. Key topics include representations, event cameras in field robotics, opportunities compared to conventional image sensors, hardware, and software architectures for neuromorphic processing, event-based simultaneous localization and mapping (SLAM). We will emphasize both academic progress and industrial applications, with a focus on challenges, opportunities, and lessons from experimentation or field deployment. Attendees will gain insight into current capabilities and limitations of the technology, as well as open challenges in the field. We aim to make the workshop a forum for roboticists to gain exposure to the technology and apply it to their research domains.