Blending Controllers via Multi-Objective Bandits

Gohari, Parham; Djeumou, Franck; Vinod, Abraham P.; Topcu, Ufuk

doi:10.48550/arxiv.2007.15755

Cited by 2 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The decision-maker must therefore select a controller, which in turn, selects the parameters. Typically, a performant controller, 𝐶 𝑝 , is used to ensure that the autonomous CPS focuses on its performance objectives (e.g., minimize the travel time) [12]. Performant controllers are often designed using ML-based approaches and trained on data that closely mimics the operating conditions.…”

Section: Problem Setup and Model 21 Problem Setupmentioning

confidence: 99%

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Luo¹,

Ramakrishna²,

Pettet³

et al. 2023

Preprint

View full text Add to dashboard Cite

Learning Enabled Components (LEC) have greatly assisted cyberphysical systems in achieving higher levels of autonomy. However, LEC's susceptibility to dynamic and uncertain operating conditions is a critical challenge for the safety of these systems. Redundant controller architectures have been widely adopted for safety assurance in such contexts. These architectures augment LEC "performant" controllers that are difficult to verify with "safety" controllers and the decision logic to switch between them. While these architectures ensure safety, we point out two limitations. First, they are trained offline to learn a conservative policy of always selecting a controller that maintains the system's safety, which limits the system's adaptability to dynamic and non-stationary environments. Second, they do not support reverse switching from the safety controller to the performant controller, even when the threat to safety is no longer present. To address these limitations, we propose a dynamic simplex strategy with an online controller switching logic that allows two-way switching. We consider switching as a sequential decision-making problem and model it as a semi-Markov decision process. We leverage a combination of a myopic selector using surrogate models (for the forward switch) and a non-myopic planner (for the reverse switch) to balance safety and performance. We evaluate this approach using an autonomous vehicle case study in the CARLA simulator using different driving conditions, locations, and component failures. We show that the proposed approach results in fewer collisions and higher performance than state-ofthe-art alternatives.

show abstract

Section: Problem Setup and Model 21 Problem Setupmentioning

confidence: 99%

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Luo¹,

Ramakrishna²,

Pettet³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In the context of IPP, Ma et al have formulated a "softblending" function [8] in which each candidate location is evaluated with weighted summation of the travel distance and GPR's estimation variance to balance between the final performance in precision and efficiency. In [10], a more relevant model, outside IPP, was built with multi-armed bandits upon two types of specialised robotic controllersone of high performance and one of high safety -for mobile robots to best perform the collision avoidance in autonomous navigation. The main distinction in our work is that the lowlevel specialised controllers are not the expensive products of RL, and instead our goal is to combine relatively cheap planners to produce a sophisticated, complementary plan that could not be generated by simply using any of them alone.…”

Section: B Blending Controllersmentioning

confidence: 99%

Adaptive Selection of Informative Path Planning Strategies via Reinforcement Learning

Choi¹,

Cielniak²

2021

Preprint

View full text Add to dashboard Cite

In our previous work, we designed a systematic policy to prioritize sampling locations to lead significant accuracy improvement in spatial interpolation by using the prediction uncertainty of Gaussian Process Regression (GPR) as "attraction force" to deployed robots in path planning. Although the integration with Traveling Salesman Problem (TSP) solvers was also shown to produce relatively short travel distance, we here hypothesise several factors that could decrease the overall prediction precision as well because sub-optimal locations may eventually be included in their paths. To address this issue, in this paper, we first explore "local planning" approaches adopting various spatial ranges within which next sampling locations are prioritized to investigate their effects on the prediction performance as well as incurred travel distance. Also, Reinforcement Learning (RL)-based high-level controllers are trained to adaptively produce blended plans from a particular set of local planners to inherit unique strengths from that selection depending on latest prediction states. Our experiments on use cases of temperature monitoring robots demonstrate that the dynamic mixtures of planners can not only generate sophisticated, informative plans that a single planner could not create alone but also ensure significantly reduced travel distances at no cost of prediction reliability without any assist of additional modules for shortest path calculation.

show abstract

Blending Controllers via Multi-Objective Bandits

Abstract: Equal contributionPreprint. Under review.

Cited by 2 publications

References 19 publications

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Dynamic Simplex: Balancing Safety and Performance in Autonomous Cyber Physical Systems

Adaptive Selection of Informative Path Planning Strategies via Reinforcement Learning

Contact Info

Product

Resources

About