Multi-Agent Dynamic Ergodic Search with Low-Information Sensors

Coffin, Howard; Abraham, Ian; Sartoretti, Guillaume; Dillstrom, Tyler; Choset, Howie

doi:10.1109/icra46639.2022.9812037

Cited by 8 publications

(14 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Novel work on ergodic search methods has emerged to compute continuous coverage trajectories of an area given enough time [16]. Ergodic search methods optimize robot trajectories against some underlying distributed information over an area which the robot can explore [24,22,18,3]. The success of ergodic search methods compared to prior methods is attributed to the unique ergodic metric used in the trajectory optimization.…”

Section: A Coverage-based and Ergodic Search Methodsmentioning

confidence: 99%

“…Methods that do consider time will often do so in bi-level optimization or as hybrid approaches that still require some form of node-based discretization [14,15]. In contrast, recent advances in ergodic coverage-based search methods have demonstrated it is possible to consider time more explicitly in autonomous coverage problems [16,17,18,19,20,21].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Time Optimal Ergodic Search

Dong,

Berger,

Abraham

2023

Robotics: Science and Systems XIX

View full text Add to dashboard Cite

Robots with the ability to balance time against the thoroughness of search have the potential to provide time-critical assistance in applications such as search and rescue. Current advances in ergodic coverage-based search methods have enabled robots to completely explore and search an area in a fixed amount of time. However, optimizing time against the quality of autonomous ergodic search has yet to be demonstrated. In this paper, we investigate solutions to the time-optimal ergodic search problem for fast and adaptive robotic search and exploration. We pose the problem as a minimum time problem with an ergodic inequality constraint whose upper bound regulates and balances the granularity of search against time. Solutions to the problem are presented analytically using Pontryagin's conditions of optimality and demonstrated numerically through a direct transcription optimization approach. We show the efficacy of the approach in generating time-optimal ergodic search trajectories in simulation and with drone experiments in a cluttered environment. Obstacle avoidance is shown to be readily integrated into our formulation, and we perform ablation studies that investigate parameter dependence on optimized time and trajectory sensitivity for search.

show abstract

Section: A Coverage-based and Ergodic Search Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Time Optimal Ergodic Search

Dong,

Berger,

Abraham

2023

Robotics: Science and Systems XIX

View full text Add to dashboard Cite

show abstract

“…In order to locate the targets, the agent takes a measurement at time t at/around its current location x t ∈ E. The agent is equipped with an accurate binary sensor, which output measurements z i,t ∈ {0, 1} indicating the presence of target i in the sensor's FoV/footprint S(x t ) ⊆ E [5]. In our context, we assume a circular sensor footprint with radius 0.1, and consider a 2D unit square as search domain E = [0, 1] 2 .…”

Section: B Sensor Setupmentioning

confidence: 99%

“…For each training batch, the number of nodes in the graph is uniformly randomized within [100, 200] with 10 nearest neighboring nodes connected. The range of history sequence length is randomly drawn within [50, 100], and the number of targets in [2,5]. While persistent monitoring is an infinite-horizoned problem, we terminate each episode once it exceeds 256 decision steps to allow for episodic training.…”

Section: B Network Structurementioning

confidence: 99%

“…In this paper, we specifically investigate the problem of controlling a mobile robot to persistently monitor a set of mobile targets moving over a given domain. Our agent is equipped with a binary sensor with limited Field-of-View (FoV) [5] (e.g., a camera with a visual classifier for target detection), and may or may not know the number and initial positions of the targets a priori. The underlying motion model/dynamics of the targets is assumed unknown.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatio-Temporal Self-Attention Network for Video Saliency Prediction

Wang

et al. 2023

IEEE Trans. Multimedia

View full text Add to dashboard Cite

This work focuses on the persistent monitoring problem, where a set of targets moving based on an unknown model must be monitored by an autonomous mobile robot with a limited sensing range. To keep each target's position estimate as accurate as possible, the robot needs to adaptively plan its path to (re-)visit all the targets and update its belief from measurements collected along the way. In doing so, the main challenge is to strike a balance between exploitation, i.e., re-visiting previously-located targets, and exploration, i.e., finding new targets or re-acquiring lost ones. Encouraged by recent advances in deep reinforcement learning, we introduce an attention-based neural solution to the persistent monitoring problem, where the agent can learn the inter-dependencies between targets, i.e., their spatial and temporal correlations, conditioned on past measurements. This endows the agent with the ability to determine which target, time, and location to attend to across multiple scales, which we show also helps relax the usual limitations of a finite target set. We experimentally demonstrate that our method outperforms other baselines in terms of number of targets visits and average estimation error in complex environments. Finally, we implement and validate our model in a drone-based simulation experiment to monitor mobile ground targets in a high-fidelity simulator.

show abstract

Persistent surveillance by heterogeneous multi-agents using mutual information based on observation capability

Kobayashi,

Higuchi

2024

Artif Life Robotics

View full text Add to dashboard Cite

Multi-Agent Dynamic Ergodic Search with Low-Information Sensors

Cited by 8 publications

References 28 publications

Time Optimal Ergodic Search

Time Optimal Ergodic Search

Spatio-Temporal Self-Attention Network for Video Saliency Prediction

Persistent surveillance by heterogeneous multi-agents using mutual information based on observation capability

Contact Info

Product

Resources

About