Title: From Uncertain Perception to Robust Action: Some Adventures with Autonomous Driving
Abstract:
A defining feature of robotics today is the use of learning and autonomy in the inner loop of systems that are actually being deployed in the real world. Autonomous driving is a prototypical example, drawing on techniques ranging from neural network models for perception to reinforcement learning of policies for action. I will begin with a brief description of the architecture of such a pipeline being developed at a startup company I am involved with.
The single biggest obstacle to scaling up such technologies from initial trials to mass market products is the problem of ensuring the safety of such systems in an open environment. We are approaching this problem through a layered architecture that combines fairly established ideas regarding risk and verification with emerging architectures for robot learning. I will describe three results that arise from this work.
I will conclude with some higher level observations regarding the business of technology, from my perspective as an academic who has tried to take research ideas into a deployable product.
Bio:
Dr. Subramanian Ramamoorthy is a Reader (Associate Professor) in the School of Informatics, University of Edinburgh, where he has been on the faculty since 2007. He is an Executive Committee Member for the Edinburgh Centre for Robotics. He received his PhD in Electrical and Computer Engineering from The University of Texas at Austin in 2007. He has been an elected Member of the Young Academy of Scotland at the Royal Society of Edinburgh, and a Visiting Professor at Stanford University and the University of Rome “La Sapienza”. He also serves as Vice President - Prediction and Planning at FiveAI, a UK-based startup company focussed on developing a technology stack for autonomous vehicles. His research focus is on robot learning and decision-making under uncertainty, with particular emphasis on achieving safe and robust autonomy in human-centered environments.