Charting the trade-off between design complexity and plan execution under probabilistic actions

Saberifar, Fatemeh Zahra; Shell, Dylan A.; O’Kane, Jason M.

doi:10.1109/icra46639.2022.9811751

Cited by 6 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ignoring power analysis and effect sizes has practical consequences. In robotics, design choices come with tradeoffs between manufacturing cost, design complexity, execution performance, computational costs, and energy consumption (Nickel et al, 2022;Saberifar et al, 2022).…”

Section: The Effects Of Motionmentioning

confidence: 99%

Biological Motion in Robots and Users’ Action Understanding

Wan,

Bennett

2024

Preprint

View full text Add to dashboard Cite

When a robot’s motion is more predictable, users experience lower stress and perceive the robot as more competent. The present study examined previously unexplored relationships between robot motion, anthropomorphic forms, and human action understanding. In an online experiment, we tracked participants’ (n = 62) gaze as they watched a robotic arm, humanoid, and human pouring coffee, either in biological (smooth, curvilinear) or nonbiological (segmented, sudden) motion. Additionally, the action outcome was either correct (filling a cup) or erroneous (spilling); participants gave key-press responses to predict which outcome. Motion had significant main effects on both predictive gaze and error detection response times. However, while predictive gaze was statistically equivalent when observing the robotic arm and humanoid in biological motion, it was counterintuitively lower when observing humans. Taken together, our findings contribute evidence to support an expanded role of predictable motion in robot design.

show abstract

Section: The Effects Of Motionmentioning

confidence: 99%

Biological Motion in Robots and Users’ Action Understanding

Wan,

Bennett

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Importantly, methodologies and representations will need to scale in order to provide decision support for large and complex systems because field robotic systems oftentimes have many subsystems, components, parameters, and permutations of feasible configurations; a probabilistic framework will likely provide the basis for stages 2 through 5. Recent literature on how roboticists can select sensors (Rahmani, Shell, and O'Kane 2021) and explore trade-offs in the design space when building robots (Saberifar, Shell, and O'Kane 2022) may provide useful insights into understanding and modeling systems.…”

Section: Future Directionsmentioning

confidence: 99%

Taxonomy of A Decision Support System for Adaptive Experimental Design in Field Robotics

Gregory¹,

Al-Hussaini²,

Agha–mohammadi³

et al. 2022

Preprint

View full text Add to dashboard Cite

Experimental design in field robotics is an adaptive humanin-the-loop decision-making process in which an experimenter learns about system performance and limitations through interactions with a robot in the form of constructed experiments. This can be challenging because of system complexity, the need to operate in unstructured environments, and the competing objectives of maximizing information gain while simultaneously minimizing experimental costs. Based on the successes in other domains, we propose the use of a Decision Support System (DSS) to amplify the human's decision-making abilities, overcome their inherent shortcomings, and enable principled decision-making in field experiments. In this work, we propose common terminology and a six-stage taxonomy of DSSs specifically for adaptive experimental design of more informative tests and reduced experimental costs. We construct and present our taxonomy using examples and trends from DSS literature, including works involving artificial intelligence and Intelligent DSSs. Finally, we identify critical technical gaps and opportunities for future research to direct the scientific community in the pursuit of next-generation DSSs for experimental design.

show abstract

“…Designing robotic systems that face multiple-objectives requires representations of Pareto-fronts, which is a well-known problem in optimization [9], [20]- [23], but also studied for specific robotics applications [3], [8], [24]- [27]. A common approach to multi-objective optimization is linear scalarization, i.e., using the weighted sum of the individual objective functions to pose a single optimization problem [9].…”

Section: B Related Workmentioning

confidence: 99%

“…Task allocation and sequencing is a fundamental problem in multi-robot systems, with applications in environmental monitoring [1]- [3], service in homes and health-care facilities [4], [5], pickup-and-delivery [6] and autonomous mobility-on-demand [7], [8]. However, many real world problems pose a diverse set of requirements for robot capabilities, leading to specialized robot designs [9], [10]. Deploying a heterogenous fleet of robots offers the most efficient use of resources, yet poses new challenges due to the increased combinatorial complexity.…”

Section: Introductionmentioning

confidence: 99%

Online Multi-Robot Task Assignment with Stochastic Blockages

Wilde

Alonso–Mora

2022

2022 IEEE 61st Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

We study the problem of finding statistically distinct plans for stochastic planning and task assignment problems such as online multi-robot pickup and delivery (MRPD) when facing multiple competing objectives. In many real-world settings robot fleets do not only need to fulfil delivery requests, but also have to consider auxiliary objectives such as energy efficiency or avoiding human-centered work spaces. We pose MRPD as a multiobjective optimization problem where the goal is to find MRPD policies that yield different trade-offs between given objectives. There are two main challenges: 1) MRPD is computationally hard, which limits the number of trade-offs that can reasonably be computed, and 2) due to the random task arrivals, one needs to consider statistical variance of the objective values in addition to the average. We present an adaptive sampling algorithm that finds a set of policies which i) are approximately optimal, ii) approximate the set of all optimal solutions, and iii) are statistically distinguishable. We prove completeness and adapt a state-of-the-art MRPD solver to the multi-objective setting for three example objectives. In a series of simulation experiments we demonstrate the advantages of the proposed method compared to baseline approaches and show its robustness in a sensitivity analysis. The approach is general and could be adapted to other multi-objective task assignment and planning problems under uncertainty.

show abstract

Charting the trade-off between design complexity and plan execution under probabilistic actions

Cited by 6 publications

References 23 publications

Biological Motion in Robots and Users’ Action Understanding

Biological Motion in Robots and Users’ Action Understanding

Taxonomy of A Decision Support System for Adaptive Experimental Design in Field Robotics

Online Multi-Robot Task Assignment with Stochastic Blockages

Contact Info

Product

Resources

About