Survey of Methods for Planning Safe Routes for Autonomous Land, Water, and Aerial Vehicles and Analysis of Their Cybersecurity

Abstract

To address the challenges of ensuring the safe operation of autonomous vehicles, this paper presents a comparative evaluation of the methods developed by the author for determining safe and optimal routes in multi-vehicle cooperative and non-cooperative navigation situations with varying traffic density, taking into account collision risk assessment. For each of the three types of autonomous vehicles—land, water, and air—we propose a kinematic trajectory (KT) method, a dynamic trajectory (DT) method, and a game trajectory (GT) method. Linear and dynamic programming, artificial neural networks, and matrix and positional multiplayer games were used for this purpose. Simulation studies of these three methods for determining safe trajectories in test navigation situations of varying complexity allowed for a comparison of their effectiveness in various environmental states. Additionally, biomimetic solutions for autonomous vehicles were described for each of their three operational areas. This analysis also highlighted the types and methods of cybersecurity used for individual types of autonomous vehicles.