Belief Space Metareasoning for Exception Recovery

Svegliato, Justin; Wray, Kyle Hollins; Witwicki, Stefan J.; Biswas, Joydeep; Zilberstein, Shlomo

doi:10.1109/iros40897.2019.8967676

Cited by 20 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other planning domains have also adopted meta-reasoning, including algorithm selection in sorting [24] and belief space meta-reasoning in autonomous driving applications [25].…”

Section: Related Workmentioning

confidence: 99%

Learning When to Quit: Meta-Reasoning for Motion Planning

Sung¹,

Kaelbling²,

Lozano-Pérez³

2021

Preprint

View full text Add to dashboard Cite

Anytime motion planners are widely used in robotics. However, the relationship between their solution quality and computation time is not well understood, and thus, determining when to quit planning and start execution is unclear. In this paper, we address the problem of deciding when to stop deliberation under bounded computational capacity, so called meta-reasoning, for anytime motion planning. We propose data-driven learning methods, model-based and model-free meta-reasoning, that are applicable to different environment distributions and agnostic to the choice of anytime motion planners. As a part of the framework, we design a convolutional neural network-based optimal solution predictor that predicts the optimal path length from a given 2D workspace image. We empirically evaluate the performance of the proposed methods in simulation in comparison with baselines.

show abstract

“…Other planning domains have also adopted meta-reasoning, including algorithm selection in sorting [24] and belief space meta-reasoning in autonomous driving applications [25].…”

Section: Related Workmentioning

confidence: 99%

Learning When to Quit: Meta-Reasoning for Motion Planning

Sung¹,

Kaelbling²,

Lozano-Pérez³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…An SSP is a general-purpose model for sequential decision making in stochastic domains with an objective of finding the least-cost path from a start state to a goal state. This model has been used in a wide range of applications, including exception recovery [37], electric vehicle charging [32], search and rescue [27], and autonomous navigation [45].…”

Section: Domain Modelmentioning

confidence: 99%

Improving Competence for Reliable Autonomy

Basich

Svegliato

Wray³

et al. 2020

Electron. Proc. Theor. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

Given the complexity of real-world, unstructured domains, it is often impossible or impractical to design models that include every feature needed to handle all possible scenarios that an autonomous system may encounter. For an autonomous system to be reliable in such domains, it should have the ability to improve its competence online. In this paper, we propose a method for improving the competence of a system over the course of its deployment. We specifically focus on a class of semiautonomous systems known as competence-aware systems that model their own competence-the optimal extent of autonomy to use in any given situation-and learn this competence over time from feedback received through interactions with a human authority. Our method exploits such feedback to identify important state features missing from the system's initial model, and incorporates them into its state representation. The result is an agent that better predicts human involvement, leading to improvements in its competence and reliability, and as a result, its overall performance.

show abstract

“…As noted earlier, we focus on the CAS illustrated in Figure 1, which represents a class of CAS with four distinct levels of autonomy and four feedback signals. Recent work on autonomous vehicles [28] and autonomous mobile robots [22] suggests that this class of CAS represents a wide range of autonomous systems. In Figure 1, the policy π produces an action a and a level l for every state s. l dictates the manner in which the system carries out the action a, and the autonomy profile κ restricts the levels the π can return.…”

Section: Sample Casmentioning

confidence: 99%

“…Hence, an initial level of competence could be determined during testing and evaluation, but adjustments must be made when the system is deployed. Even when developers aim to err on the side of caution and define a lower level of autonomy as the default, it is also possible to unintentionally infer from initial testing that the system is more competent than it really is [18,22]. Therefore, developing mechanisms to explicitly represent, reason about, and adjust the level of autonomy is an important challenge in artificial intelligence.…”

Section: Introductionmentioning

confidence: 99%

Learning to Optimize Autonomy in Competence-Aware Systems

Basich¹,

Svegliato²,

Wray³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Interest in semi-autonomous systems (SAS) is growing rapidly as a paradigm to deploy autonomous systems in domains that require occasional reliance on humans. This paradigm allows service robots or autonomous vehicles to operate at varying levels of autonomy and offer safety in situations that require human judgment. We propose an introspective model of autonomy that is learned and updated online through experience and dictates the extent to which the agent can act autonomously in any given situation. We define a competence-aware system (CAS) that explicitly models its own proficiency at different levels of autonomy and the available human feedback. A CAS learns to adjust its level of autonomy based on experience to maximize overall efficiency, factoring in the cost of human assistance. We analyze the convergence properties of CAS and provide experimental results for robot delivery and autonomous driving domains that demonstrate the benefits of the approach.

show abstract

Belief Space Metareasoning for Exception Recovery

Cited by 20 publications

References 8 publications

Learning When to Quit: Meta-Reasoning for Motion Planning

Learning When to Quit: Meta-Reasoning for Motion Planning

Improving Competence for Reliable Autonomy

Learning to Optimize Autonomy in Competence-Aware Systems

Contact Info

Product

Resources

About