Trajectory Planning for Autonomous Cars in Low-Structured and Unstructured Environments: A Systematic Review

Autonomous driving research has gained attention in recent years due to its great potential in reshaping transportation systems. It is highly noticeable that most research focuses on well-maintained urban roads, whereas much less effort is pushed for low or unstructured scenarios usually found in su...

Full description

Saved in:
Bibliographic Details
Main Authors: Cristiano Souza de Oliveira, Rafael De S. Toledo, Vitor H. Tulux Oliveira Victorio, Aldo von Wangenheim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10926194/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Autonomous driving research has gained attention in recent years due to its great potential in reshaping transportation systems. It is highly noticeable that most research focuses on well-maintained urban roads, whereas much less effort is pushed for low or unstructured scenarios usually found in suburban or countryside areas of emerging countries. This study systematically reviews state-of-the-art trajectory planning for car-like vehicles in these environments. We analyzed 4601 papers published in the last 10 years and selected 75 studies. Based on the selected research, we summarize the techniques for global planning, local planning, motion control, and multi-vehicle collaborative planning. In particular, we highlight the strengths and weaknesses of the local planning approaches and discuss and identify common limitations, such as sensor faults, terrain variations, safety and regulatory risks. Moreover, we summarize the most adopted visual input sensors, their associated data structures that are used as search spaces for local planning, and sensor fusion techniques, which complement the advantages of different types of sensors. Furthermore, we provide a systematic schema that relates the dependencies between the visual sensors, data-representations, and local planning techniques. From this, we identify gaps in the current literature, suggesting possible directions for future work.
ISSN:2169-3536