Multi-UAV Search and Rescue in Wilderness Using Smart Agent-Based Probability Models

TL;DR

A smart agent-based probability model that combines Monte Carlo simulations with an agent strategy list, mimicking the behavior of a lost person in the wildness areas is proposed and a distributed Multi-UAV receding horizon search strategy with dynamic partitioning is developed.

Abstract

The application of Multiple Unmanned Aerial Vehicles (Multi-UAV) in Wilderness Search and Rescue (WiSAR) significantly enhances mission success due to their rapid coverage of search areas from high altitudes and their adaptability to complex terrains. This capability is particularly crucial because time is a critical factor in searching for a lost person in the wilderness; as time passes, survival rates decrease and the search area expands. The probability of success in such searches can be further improved if UAVs leverage terrain features to predict the lost person's position. In this paper, we aim to enhance search missions by proposing a smart agent-based probability model that combines Monte Carlo simulations with an agent strategy list, mimicking the behavior of a lost person in the wildness areas. Furthermore, we develop a distributed Multi-UAV receding horizon search strategy with dynamic partitioning, utilizing the generated probability density model as prior information to prioritize locations where the lost person is most likely to be found. Simulated search experiments across different terrains have been conducted to validate the search efficiency of the proposed methods compared to other benchmark methods.

Figures (8)

• Figure 1: Illustration of a distributed multi-UAV system in WiSAR. UAVs communicate with adjacent UAVs within their communication range. One of the predicted trajectories of the lost person are shown in green line.
• Figure 2: System architecture of the proposed framework.
• Figure 3: The guideline map is generated using a set of rays originating from LKP, shown as the green lines. A simulated agent (black point) initially selects a ray (blue line) to follow. Upon encountering an uncrossable mountain, the agent switches to an adjacent ray to find a traversable path, following the river's edge until he/she find a way out of the area.
• Figure 4: Estimated lost person probability distribution over time: The simulated agents are represented by green particles, and the environmental linear features, such as trails and roads, are shown as the white dash line.
• Figure 5: Illustration of the iterations of the receding horizon-based searching.

Theorems & Definitions (2)

• Remark 1
• Remark 2