Learning to Rapidly Re-Contact the Lost Plume in Chemical Plume Tracing
نویسندگان
چکیده
Maintaining contact between the robot and plume is significant in chemical plume tracing (CPT). In the time immediately following the loss of chemical detection during the process of CPT, Track-Out activities bias the robot heading relative to the upwind direction, expecting to rapidly re-contact the plume. To determine the bias angle used in the Track-Out activity, we propose an online instance-based reinforcement learning method, namely virtual trail following (VTF). In VTF, action-value is generalized from recently stored instances of successful Track-Out activities. We also propose a collaborative VTF (cVTF) method, in which multiple robots store their own instances, and learn from the stored instances, in the same database. The proposed VTF and cVTF methods are compared with biased upwind surge (BUS) method, in which all Track-Out activities utilize an offline optimized universal bias angle, in an indoor environment with three different airflow fields. With respect to our experimental conditions, VTF and cVTF show stronger adaptability to different airflow environments than BUS, and furthermore, cVTF yields higher success rates and time-efficiencies than VTF.
منابع مشابه
Chemical Plume Tracing by Discrete Fourier Analysis and Particle Swarm Optimization
A novel methodology for solving the chemical plume tracing problem that utilizes data from a network of stationary sensors has been developed in this study. During a toxic chemical release and dispersion incident, the imperative need of first responders is to determine the physical location of the source of chemical release in the shortest possible time. However, the chemical plume that develop...
متن کاملChemical Plume Tracing and Mapping via Swarm Robots
This chapter addresses the key issues of chemical plume mapping and tracing via swarm robots. First, the authors present the models of turbulent odor plumes with both non-buoyant and buoyant features, which can efficiently evaluate strategies for tracing plumes, identifying their sources in two or threedimensions. Second, the authors use the Monte Carlo technique to optimize moth-inspired plume...
متن کاملTracking of Fluid-Advected Odor Plumes: Strategies Inspired by Insect Orientation to Pheromone
Autonomous vehicles with plume tracing capabilities would be valuable for finding chemical sources in fluid flows. This article considers strategies allowing autonomous vehicles to find and trace an odor plume to its source. These strategies are inspired by the maneuvers of moths flying upwind along a pheromone plume. Although moth maneuvers are well documented, the mechanisms underlying sensor...
متن کاملPlume mapping via hidden Markov methods
This paper addresses the problem of mapping likely locations of a chemical source using an autonomous vehicle operating in a fluid flow. The paper reviews biological plume-tracing concepts, reviews previous strategies for vehicle-based plume tracing, and presents a new plume mapping approach based on hidden Markov methods (HMM). HMM provide efficient algorithms for predicting the likelihood of ...
متن کاملFormation Control of Multiple AUVs for Moth-Inspired Plume Tracing
This paper presents a novel strategy for multiple autonomous underwater vehicles (AUVs) to trace a chemical plume and to identify its source location in a simulated fluid-advected environment with scales of 100 meters. First, we develop a coordination mechanism for selecting a leader among multiple AUVs which performs moth-inspired plume tracing. Second, we use a formation algorithm to control ...
متن کامل