Predictive Probability Density Mapping for Search and Rescue Using An Agent-Based Approach with Sparse Data: Simulation Data

Predicting the location where a lost person could be found is crucial for    search and rescue operations with limited resources.    To improve the    precision and efficiency of these predictions,    simulated agents can be    created to emulate the behavior of the lost person.    Within this study,     we introduce an innovative agent-based model designed    to replicate diverse psychological profiles of lost persons,     allowing these agents to navigate real-world landscapes while making    decisions autonomously     without the need for location-specific training.     The probability distribution map depicting the potential location of the lost person    emerges through a combination of Monte Carlo simulations and    mobility-time-based sampling.     Validation of the model is achieved using real-world Search and Rescue data to train a Gaussian Process model.     This allows generalization of the data to sample initial starting points for the agents during validation.    Comparative analysis with historical data    showcases promising outcomes relative to alternative methods.     This work introduces a flexible agent that can be employed in search and rescue operations,     offering adaptability across various geographical locations.