Informed Rapidly Exploring Random Trees for Autonomous Mining Vehicles

Thumbnail Image

Files

master_Caetano_Ezequiel_Ricardo_2024.pdf (9.05 MB)

URL

Journal Title

Journal ISSN

Volume Title

Sähkötekniikan korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2024-08-19

Department

Major/Subject

Autonomous Systems

Mcode

ELEC3055

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

77

Series

Abstract

This thesis explores novel methodologies in informed sampling for nonholonomic path planning in complex environments. Traditional path planning algorithms face challenges when navigating through unknown areas, requiring innovative approaches to balance computational efficiency and solution optimality. The study introduces the Heuristic Cost Rejection Sampling (HCRS) method, which adapts rejection sampling with a heuristic cost function tailored for nonholonomic constraints. The approach addresses the limitations of traditional methods by integrating penalizers for path curvature and manoeuvre complexity into the cost function, enhancing comparability and effectiveness in path planning tasks. Furthermore, the thesis investigates the Window Sampling strategy, inspired by SWIRRT*, to optimize informed sampling efficiency in intricate environments. This strategy involves dynamically sampling nodes within sliding windows along the planned path, balancing exploration and exploitation to refine path solutions incrementally. Comparative evaluations between different sampling methods highlight HCRS's superior performance in providing consistent, fast solutions with reduced solution cost variability. Experimental results across homotopy scenarios demonstrate HCRS's effectiveness in achieving faster time-to-goal and maintaining solution stability compared to traditional approaches. The study also evaluates the impact of informed node classification on solution quality and computational efficiency, revealing insights into optimizing informed sampling strategies for autonomous navigation systems. In conclusion, this thesis contributes advancements in informed sampling methodologies tailored for nonholonomic path planning, offering practical insights and performance benchmarks essential for future developments in autonomous vehicle navigation and robotics applications.

Description

Supervisor

Del Prete, Andrea

Thesis advisor

Ismail, Ofa
Le, Tuan

Keywords

informed RRT*, path planning, nonholonomic systems, mining trucks

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202412298132

Collections

[dipl] Sähkötekniikan korkeakoulu / ELEC