Stanford Seminar - Robots in Dynamic Tasks: Learning, Risk, and Safety

March 10, 2023
Joel Burdick of Caltech

Autonomous robots are increasing applied to tasks that involve complex maneuvers and dynamic environments that are difficult to model a priori. Various types of learning methods have been proposed to fill this modeling gap. To motivate the need for learning complex fluid-structure interactions, we first review the SQUID (a ballistically launched and self-stabilizing drone) and PARSEC (an aerial manipulator that can deliver self-anchoring sensor network modules) systems. Next we show how to learn basic fluid-structure interactions using Koopman spectral techniques, and incorporate the learned model into a real-time nonlinear model predictive control framework. The performance of this approach is demonstrated on small drones that operate very close to the ground, where the ground effect normally destabilizes flight. Operational risk abounds in complex robotic tasks. This risk arises both from the uncertain environment, and from incompletely learned models. After reviewing the basics of coherent risk measures, we will show how simple risk aware terrain analysis improved the performance of our legged and wheeled robots in the DARPA Subterranean challenge. Then we will introduce an on-line method to learn the dynamics of an apriori unknown dynamical obstacle, and robustly avoid the obstacle using a novel risk-based, distributionally robust, chance constraints derived from the evolving learned model. We then introduce the concept of risk surfaces to enable fast on-line learning of a priori unknown dynamical disturbances, and show how this approach can adapt a drone to wind disturbances with only 45 seconds of on-line data gathering.

0:00 Introduction
2:47 SQUID I: Key Design Elements
4:00 SQUID II: Vision-based Autonomous Stabilization
6:22 Planetary Exploration Applications
8:01 PARSEC: Payload Anchoring Robotic System for the Exploration of Cliffs Task Motivation and Description
8:32 PARSEC: Aerial Manipulator
9:23 Deployment Interface and Payload Design
10:06 Mission Architecture for Autonomous Deployment
11:39 But what about the real world?
12:30 Machine Learning & Nonlinear Vehicle Control
17:24 Using learned lifted bilinear models for nonlinear MPC
20:27 Learning quadrotor dynamics to improve close-to-ground trajectory tracking
21:00 Learning quadrotor dynamics to improve close-to- ground trajectory tracking performance
24:00 Planning under uncertainty
24:56 Risk-Aware Planning: Chance Constraints
28:15 The DARPA Subterranean Challenge
29:05 STEP: Stochastic Traversability Evaluation and Planning
34:07 Risk-Aware Avoidance of Unknown Dynamic Ostacles
42:54 Robust Risk-Based Learning of Disturbances
46:41 Learning and Introspective Control (LINC)

[vivekacharya]

stanford is so much involved in robotics...