"Drunk Man" Saves Our Lives: Route Planning by a Biased Random Walk Mode

TL;DR

This paper introduces 'DroneGo', a disaster response system using a drone fleet and a novel route planning method combining genetic algorithms with a biased random walk model inspired by a drunk man, demonstrating high performance in complex terrains.

Contribution

It presents a new route planning approach that integrates genetic algorithms with a biased random walk model for disaster response, inspired by human-like stochastic exploration.

Findings

High performance in route planning achieved

Effective exploration of feasible routes in complex terrains

Combines stochasticity with objective biasing

Abstract

Based on the hurricane striking Puerto Rico in 2017, we developed a transportable disaster response system "DroneGo" featuring a drone fleet capable of delivering the medical package and videoing roads. Covering with a genetic algorithm and a biased random walk model mimicking a drunk man to explore feasible routes on a field with altitude and road information. A proposal mechanism guaranteeing stochasticity and an objective function biasing randomness are combined. The results showed high performance though time-consuming.