Physics-Based Traffic Motion Techniques In Motion Planning Of Intelligent Autonomous Robots

Abstract

We provide a motion strategic plan for an automated robot navigation pathway that takes into account both the ambiguity posed by the automated robot as well as the uncertainty posed by other vehicles involved. The ambiguity along the anticipated trajectory is estimated using Gaussian dispersion, as well as the future movement of vehicles involved is forecasted using such a regional planner. The ambiguity from relocation and handling is calculated for the mobile robot using Linear-Quadratic Gaussian (LQG) architecture. Due to direct ambiguity data feedback to the designer, our architecture helps a planner to prevent risky scenarios more effectively than existing safety evaluation techniques. In comparison to planning just using an LQG architecture, we show that our planner may yield safer paths.