Self-Assessment and Resilience

Long-term autonomous operation requires robots to be aware of their own performance and detect faults that might impede their goals. In this project, we are developing introspective algorithms for performance assessing under uncertainty, as well as fault detection and recovery in order to enable robots to become more resilient to perturbations in their environment.

A Framework for Robot Self-Assessment of Expected Task Performance (2022)

Tyler Frasca and Matthias Scheutz

We propose a self-assessment framework which enables a robot to estimate how well it will perform a known or novel task. The robot simulates the task by sampling performance distributions to generate a state distribution of possible outcomes and determines 1) the likelihood of success, 2) the most probable failure location, and 3) the…

Keywords: self-assessment, diarc, Human-Robot Interaction

Autonomy Reconsidered: Toward Developing Multi-Agent Systems (2021)

Michael Goodrich and Julie Adams and Matthias Scheutz

An agent’s autonomy can be viewed as the set of physically and computationally grounded algorithms that can be performed by the agent. This view leads to two useful notions related to autonomy: behav- ior potential and success potential, which can be used to measure of how well an agent fulfills its potential, call fulfillment.

Evaluating Task-General Resilience Mechanisms in a Multi-Robot Team Task (2021)

James Staley and Matthias Scheutz

Real-word intelligent agents must be able to detect sudden and unexpected changes to their task environment and effectively respond to those changes in order to function properly in the long term. We thus isolate a set of perturbations that agents ought to address and demonstrate how task-agnostic perturbation detection and mitigation…

"Can you Do This?" Self-Assessment Dialogues with Autonomous Robots Before, During, and After a Mission (2020)

Tyler Frasca and Ravenna Thielstrom and Evan Krause and Matthias Scheutz

The sophisticated capabilities of autonomous robots can be difficult for humans to understand the robot's expected performance. In this papaer, we present a robot self assessment framework which enables a robot to assess its expected task performance before, during, and after execution. We demonstrate the framework in the DIARC cognitive-robotic architecture.

Keywords: Robot-Self Task Performance Assessment, Human-Robot Interaction

Going Cognitive: A Demonstration of the Utility of Task-General Cognitive Architectures for Adaptive Robotic Task Performance (2020)

Tyler Frasca and Zhao Han and Jordan Allspaw and Holly Yanco and Matthias Scheutz

A main advantage of cognitive architectures over specialized robotic architectures is that they are task-general and can learn new tasks online. In this paper, we provide empirical evidence to for this claim, by directly comparing a high-performing custom robotic architecture developed for the standardized robotic "FetchIt!" challenge…

Keywords: Cognitive-Robotic Architectures, Human-Robot Interaction, FetchIt! Robotics Challenge