EV-Catcher: High-Speed Object Catching Using Low-latency Event-based Neural Networks
This addresses the challenge of real-time robotic perception in dynamic environments, representing an incremental improvement with specific gains in high-speed object catching.
The paper tackled the problem of catching fast-moving objects using event-based sensors by introducing a lightweight event representation and a learning-based approach, achieving an 81% success rate in catching ping-pong balls at velocities up to 13 m/s on embedded platforms.
Event-based sensors have recently drawn increasing interest in robotic perception due to their lower latency, higher dynamic range, and lower bandwidth requirements compared to standard CMOS-based imagers. These properties make them ideal tools for real-time perception tasks in highly dynamic environments. In this work, we demonstrate an application where event cameras excel: accurately estimating the impact location of fast-moving objects. We introduce a lightweight event representation called Binary Event History Image (BEHI) to encode event data at low latency, as well as a learning-based approach that allows real-time inference of a confidence-enabled control signal to the robot. To validate our approach, we present an experimental catching system in which we catch fast-flying ping-pong balls. We show that the system is capable of achieving a success rate of 81% in catching balls targeted at different locations, with a velocity of up to 13 m/s even on compute-constrained embedded platforms such as the Nvidia Jetson NX.