2013/02/04 by N. Cao, Cao, Nannan, Kian Hsiang Low +3 · 1 citation
Computer Science · Engineering · #Advanced Multi-Objective Optimization Algorithms #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Machine Learning (cs.LG) #Multiagent Systems (cs.MA) #Robotics (cs.RO) #Robotics and Sensor-Based Localization #Target Tracking and Data Fusion in Sensor Networks
paper · pdf · doi:10.48550/arxiv.1302.0723
openalex publication_date 2013/02/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
A key problem of robotic environmental sensing and monitoring is that of\nactive sensing: How can a team of robots plan the most informative observation\npaths to minimize the uncertainty in modeling and predicting an environmental\nphenomenon? This paper presents two principled approaches to efficient\ninformation-theoretic path planning based on entropy and mutual information\ncriteria for in situ active sensing of an important broad class of\nwidely-occurring environmental phenomena called anisotropic fields. Our\nproposed algorithms are novel in addressing a trade-off between active sensing\nperformance and time efficiency. An important practical consequence is that our\nalgorithms can exploit the spatial correlation structure of Gaussian\nprocess-based anisotropic fields to improve time efficiency while preserving\nnear-optimal active sensing performance. We analyze the time complexity of our\nalgorithms and prove analytically that they scale better than state-of-the-art\nalgorithms with increasing planning horizon length. We provide theoretical\nguarantees on the active sensing performance of our algorithms for a class of\nexploration tasks called transect sampling, which, in particular, can be\nimproved with longer planning time and/or lower spatial correlation along the\ntransect. Empirical evaluation on real-world anisotropic field data shows that\nour algorithms can perform better or at least as well as the state-of-the-art\nalgorithms while often incurring a few orders of magnitude less computational\ntime, even when the field conditions are less favorable.\n