This webinar was presented on Tuesday, August 18, 2020
Please click below to watch on demand
Latency plays crucial role in many robotic applications. Achieving low latency helps solve many real-world problems in SLAM, obstacle avoidance, gripping and inspection systems. Edge systems can exploit sparsity in time, connection, activation and space to achieve low latency without sacrificing power. This requires an innovative computing architecture for robotic applications. This innovative architecture also allows the combination of machine learning inference with the classical mathematical techniques required by traditional linear and nonlinear control systems.
This talk will introduce the concepts of sparsity, how it can be leveraged by robotic applications and finally an architecture that exploits it to achieve low latency.
Attendees of this webinar will:
- Understand how different compute architectures provide different capabilities for robotics
- Understand how algorithms and architectures must be linked for best performance
- Understand how to access and take advantage of all four types of sparsity
Chief Scientific Officer
GrAI Matter Labs
The Robot Report