Reliable Robotic Task Execution in the Face of Anomalies
Authors:
- Bharath SanthanamBSBharath Santhanam
- Alex MitrevskiAMAlex Mitrevski
- Santosh ThodukaSTSantosh Thoduka
- Sebastian HoubenSHSebastian Houben
- Teena HassanTHTeena Hassan
Preprint
Abstract
Learned robot policies have consistently been shown to be versatile, but they typically have no built-in mechanism for handling the complexity of open environments, making them prone to execution failures; this implies that deploying policies without the ability to recognise and react to failures may lead to unreliable and unsafe robot behaviour. In this paper, we present a framework that couples a learned policy with a method to detect visual anomalies during policy deployment and to perform recovery behaviours when necessary, thereby aiming to prevent failures. Specifically, we train an anomaly detection model using data collected during nominal executions of a trained policy. This model is then integrated into the online policy execution process, so that deviations from the nominal execution can trigger a three-level sequential recovery process that consists of (i) pausing the execution temporarily, (ii) performing a local perturbation of the robot's state, and (iii) resetting the robot to a safe state by sampling from a learned execution success model. We verify our proposed method in two different scenarios: (i) a door handle reaching task with a Kinova Gen3 arm using a policy trained in simulation and transferred to the real robot, and (ii) an object placing task with a UFactory xArm 6 using a general-purpose policy model. Our results show that integrating policy execution with anomaly detection and recovery increases the execution success rate in environments with various anomalies, such as trajectory deviations and adversarial human interventions.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- ConditionNET: Learning Preconditions and Effects for Execution MonitoringIEEE Robotics and Automation Letters, 2024
- Diffusion policy: Visuomotor policy learning via action diffusionThe International Journal of Robotics Research, 2024
- A hybrid skill parameterisation model combining symbolic and subsymbolic elements for introspective robotsRobotics and Autonomous Systems, 2022
- CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation TasksIEEE Robotics and Automation Letters, 2022
- Learning where to trust unreliable models in an unstructured world for deformable object manipulationScience Robotics, 2021
- Recovery RL: Safe Reinforcement Learning With Learned Recovery ZonesIEEE Robotics and Automation Letters, 2021
- Recent Advances in Robot Learning from DemonstrationAnnual Review of Control, Robotics, and Autonomous Systems, 2020
- From Skills to Symbols: Learning Symbolic Representations for Abstract High-Level PlanningJournal of Artificial Intelligence Research, 2018