Obstacle-aware Adaptive Informative Path Planning for UAV-based Target Search

Abstract

Target search with unmanned aerial vehicles (UAVs) is relevant problem to many scenarios, e.g., search and rescue (SaR). However, a key challenge is planning paths for maximal search efficiency given flight time constraints. To address this, we propose the Obstacle-aware Adaptive Informative Path Planning (OA-IPP) algorithm for target search in cluttered environments using UAVs. Our approach leverages a layered planning strategy using a Gaussian Process (GP)based model of target occupancy to generate informative paths in continuous 3D space. Within this framework, we introduce an adaptive replanning scheme which allows us to trade off between information gain, field coverage, sensor performance, and collision avoidance for efficient target detection. Extensive simulations show that our OA-IPP method performs better than state-of-the-art planners, and we demonstrate its application in a realistic urban SaR scenario.

Active Sensing for Search and Tracking: A Review

“…Randomization allows to solve problems that are not solvable by deterministic algorithms [89]; in spite of this, random search is not always efficient and usually serves as a lower bound when the objective is to minimize the expected time until target detection [46]. Probabilistic Target Search (PTS) [46], [55], [95] accounts for target motion and sensing uncertainties [55] and formulates the search task within Bayesian probabilistic frameworks (e.g., Recursive Bayesian Estimation -RBE [96]- [98], Bayesian Optimization -BO [70], [99]- [101]). In this way, it is possible to encode and keep updated the knowledge about potential target locations as a probability distribution, also referred to as belief or probabilistic map [58]; this is done by treating no-detection observations (i.e., measurements with no information on target position) as negative likelihood [102].…”

“…Some APE problems are formulated as environmental mapping tasks, where the target location is inferred from the estimated spatial distribution of a physical quantity (e.g., source signal strength [16], [74]) or from occupancy maps [70], [127]. In this framework, a common approach is to tesselate the workspace Π into a grid map and apply RBE to each cell of the grid [127].…”

“…In this latter case, visual servoing is referred as active vision [5], [6], [51]. To avoid inefficient search strategies, characterized by local minima and unstable behaviors [83], POD modeling is a critical aspect when dealing with visionbased systems [70], [84]. The fundamental aspects every realistic camera observation model should consider are image edge effects, detection depth range, and scale effects: the image plane has poor detection capabilities on the edges, due to distortions caused by the camera sensor [38]; distant objects are seen at low resolution and their detection probability is low [52], [70], [95]; finally, objects too close to the camera may not be recognized, due to poor scale-invariance properties of some computer vision algorithms.…”

Acceleration-Aware Path Planning with Waypoints

“…This type of problem is well studied in literature and is termed as a minimum time search problem (MTS) [2][3][4][5][6][7][8][9][10][11][12][13][14]. The most prominent objective in these approaches is to optimize the expected time of target detection [3][4][5][6]; however, other alternative approaches involve optimizing the probability of target detection [7][8][9]15], minimizing its counterpart, i.e., probability of nondetection [10,11] or maximizing the information gain [12,13,16]. Various sub-optimal and heuristics-based algorithms such as gradient-based approaches [7,[10][11][12]15], cross-entropy optimization [2,5], Bayesian optimization algorithms [4], ant colony optimization [6], or genetic algorithms [3] have been proposed to address the NP-hard complex problem [13].…”

Learning control for transmission and navigation with a mobile robot under unknown communication rates

“…This problem is often solved dynamically, to obtain so-called informative path planning, see e.g. Meera et al (2019) and Viseras et al (2016); Popovic et al (2018) provide a good overview of this field. Closer to the present work, Fink and Kumar (2010); Penumarthi et al (2017) learn a radio map with multiple mobile robots.…”