Erik Nelson (Nuro) Robust Sensing and Perception for Self Driving Vehicles

Wednesday 28 Jul 2021, 12:00 → 13:00 US/Pacific

zoom link https://lbnl.zoom.us/j/97341258188

Abstract: 

Self-Driving Vehicles (SDVs) are beginning to emerge from research laboratories and closed testing facilities onto our public roads with limited to no supervision, where they must safely interact with human road participants. In this talk we will discuss the general anatomy of an SDV's software system, focusing towards perception submodules that interact with raw sensor inputs. We will discuss methods for hardening and robustifying these perception systems to be capable of safely handling the many unexpected situations that can arise on the road.

Short bio: 

Erik is a staff software engineer and technical lead at Nuro, where he focuses on generating and localizing within large 3D maps for autonomous driving. As an early member of the now ~1000 employee company, he has written code that touches most parts of a SDV's onboard autonomy system. Erik obtained an MS in robotics from Carnegie Mellon University, and is a PhD student in UC Berkeley's EECS department.