(Fig.1: Taxonomy of autonomous ground robots in unstructured environments.)
Research on autonomous ground robots in unstructured outdoor environments is less advanced than in structured urban settings due to challenges like environmental diversities and scene complexity. These environments-such as rural areas and rugged terrains-pose unique obstacles that are not common in structured urban areas. Despite these difficulties, autonomous ground robots in unstructured outdoor environments is crucial for applications in agriculture, mining, and military operations. Our survey reviews over 250 papers for autonomous driving in unstructured outdoor environments, covering offline mapping, pose estimation, environmental perception, path planning, end-to-end autonomous driving, datasets, and relevant challenges. We also discuss emerging trends and future research directions. This review aims to consolidate knowledge and encourage further research for autonomous ground robots in unstructured environments.
- Survey Papers
- Datasets
- Offline Mapping
- Pose Estimation
- Perception
- Path Planning
- End-to-End Driving
- Citation
(Fig.2: Inherent characteristics of unstructured environments.)
(Fig.3: Typical components of the autonomous driving system.)
(Fig.4: Challenges autonomous driving systems face in unstructured environments.)
- Diverse Odometry / SLAM
- Place Recognition & Re-Localization
| Year | Journal/Conference | Authors | Title | Task | Github |
|---|---|---|---|---|---|
| 2022 | Sensors | STV-SC: Segmentation and Temporal Verification Enhanced Scan Context for Place Recognition in Unstructured Environment | Place Recognition | - | |
| 2023 | RA-L | Place recognition of large-scale unstructured orchards with attention score maps | Place Recognition | - | |
| 2023 | IROS | Deep robust multi-robot re-localisation in natural environments | Re-Localization | - | |
| 2024 | TRO | HKU | BTC: A Binary and Triangle Combined Descriptor for 3D Place Recognition | Place Recognition | https://github.com/hku-mars/btc_descriptor |
| 2024 | arXiv | Towards Long-term Robotics in the Wild | Place Recognition | - | |
| 2024 | arXiv | Evaluation and Deployment of LiDAR-based Place Recognition in Dense Forests | Place Recognition | https://ori-drs.github.io/datasets/ | |
| 2024 | arXiv | PointNetPGAP-SLC: A 3D LiDAR-based Place Recognition Approach with Segment-level Consistency Training for Mobile Robots in Horticulture | Place Recognition | https://github.com/Cybonic/PointNetPGAP-SLC.git |
- Localization with Lightweight Map
| Year | Journal/Conference | Authors | Title | Task | Github |
|---|---|---|---|---|---|
| 2019 | ITSC | Unstructured road slam using map predictive road tracking | SLAM | - | |
| 2019 | RA-L | MIT | Maplite: Autonomous intersection navigation without a detailed prior map | localization and navigation | - |
| 2022 | TAES | HEU | TOM-odometry: A generalized localization framework based on topological map and odometry | Localization | - |
| 2023 | IROS | Global localization in unstructured environments using semantic object maps built from various viewpoints | Loc with Semantic Map | - | |
| 2023 | arXiv | ALT-Pilot: Autonomous navigation with Language augmented Topometric maps | - |
| Year | Journal/Conference | Authors | Title | Task List |
|---|---|---|---|---|
| 2021 | Advanced Robotics | Keio University | Energy-aware trajectory planning for planetary rovers | Dijkstra’s algorithm |
| 2018 | Journal of Dynamic Systems, Measurement, and Control | Ford Motor Company | A hierarchical route guidance framework for off-road connected vehicles | Dynamic Programming |
| 1994 | IEEE International Conference on Robotics and Automation | Carnegie Mellon University | Optimal and efficient path planning for partially-known environments | Dynamic A* |
| 2012 | IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology | Beijing Institute of Technology | Ara*+: Improved path planning algorithm based on ara | Anytime Repairing A* |
| 2005 | ICAPS | Carnegie Mellon University | Anytime dynamic a*: An anytime, replanning algorithm | Anytime Dynamic A* |
| 2019 | International Conference on Control, Automation and Robotics (ICCAR) | Panasonic Automotive Systems Europe GmbH | Guided hybrid a-star path planning algorithm for valet parking applications | Hybrid A* |
| 2008 | Journal of field Robotics | Stanford Artificial Intelligence Laboratory | Junior: The Stanford entry in the urban challenge | AD* |
| 2021 | ISPRS International Journal of Geo-Information | Shanghai Ocean University | Improved a-star algorithm for long-distance off-road path planning using terrain data map | improved A* algorithm |
| 2008 | IEEE intelligent vehicles symposium | Atlatec GmbH | Navigating car-like robots in unstructured environments using an obstacle sensitive cost function | A* algorithm, Voronoi cost function |
| 2022 | Machines | Chinese Academy of Sciences | A global path planning method for unmanned ground vehicles in off-road environments based on mobility prediction | PRM, improved A* algorithm |
| 2024 | IEEE Transactions on Intelligent Transportation Systems | Hunan University | Multi-vehicle collaborative trajectory planning in unstructured conflict areas based on v-hybrid a | Velocity Hybrid A* |
| 2024 | ISPRS International Journal of Geo-Information | National University of Defense Technology, China | Two-stage path planning for long-distance off-road path planning based on terrain data | PRM, the A* algorithm |
| 2010 | IEEE Transactions on Robotics | NANO-D group, INRIA | Sampling-based path planning on configuration-space costmaps | T-RRT |
| 2023 | Expert Systems with Applications | Tsinghua University | Driving risk-aversive motion planning in off-road environment | potential field-based RRT* |
| 2021 | IEEE Transactions on Automation Science and Engineering | University of Michigan–Dearborn | R2-rrt*: Reliability-based robust mission planning of off-road autonomous ground vehicle under uncertain terrain environment | R2-RRT* |
| 2023 | IEEE Transactions on Intelligent Transportation Systems | Wuhan University of Technology | Efficient reliability-based path planning of off-road autonomous ground vehicles through the coupling of surrogate modeling and rrt | ER-RRT*, surrogate modeling |
| 2024 | arXiv preprint | Tsinghua University | A risk-aware planning framework of ugvs in off-road environment | APF, Coarse2fine A* |
| 2024 | Traffic injury prevention | Suzhou Institute of construction & communications | Risk field modeling of urban tunnel based on apf | APF |
| 2020 | IEEE access | Chinese Academy of Sciences | Path planning method with improved artificial potential field—a reinforcement learning perspective | improved black hole potential fields, RL |
| 2024 | International Conference on Automation, Robotics and Applications (ICARA) | Sun Yat-sen University | On hierarchical path planning based on deep reinforcement learning in off-road environments | DWA, improved D* Lite algorithm |
| 2023 | ICARA | Politecnico di Torino | Rl-dwa omnidirectional motion planning for person following in domestic assistance and monitoring | DWA, DRL |
| 2022 | IEEE Transactions on Transportation Electrification | University of Science and Technology Beijing | Automatic parking path planning of tracked vehicle based on improved a* and dwa algorithms | DWA, improved A* algorithm |
| 2020 | International Journal of Advanced Robotic Systems | East China University of Science and Technology | Path planning of lunar robot based on dynamic adaptive ant colony algorithm and obstacle avoidance | ACO, APF |
| 2018 | Wireless Personal Communications | Army Engineering University, China | Off-road path planning based on improved ant colony algorithm | ACO |
| 2021 | Mathematical Problems in Engineering | Wuhan University of Technology | Multiobjective optimization of an off-road vehicle suspension parameter through a genetic algorithm based on the particle swarm optimization | GA, PSO |
| 2023 | International Journal of Modeling, Simulation, and Scientific Computing | Anhui Polytechnic University | Research on path planning of mobile robot based on improved genetic algorithm | improved GA |
| 2022 | arXiv preprint | University of Guelph | A novel knowledge-based genetic algorithm for robot path planning in complex environments | knowledge-based GA |
| 2022 | arXiv preprint | Samsung Advanced Institute of Technology | Vision-based autonomous driving for unstructured environments using imitation learning | Imitation Learning |
| 2023 | IEEE Transactions on Vehicular Technology | Beihang University | Trajectory planning for autonomous driving in unstructured scenarios based on deep learning and quadratic optimization | Deep Learning, Quadratic Optimization |
| 2024 | Robotics | Transilvania University of Brasov | A vision dynamics learning approach to robotic navigation in unstructured environments | RNN, DWA |
| 2018 | Proceedings of the European Conference on Computer Vision (ECCV) Workshops | Microsoft | Learning driving behaviors for automated cars in unstructured environments | DDPG |
| 2023 | IEEE Transactions on Intelligent Vehicles | Xidian University | Deep reinforcement learning-based off-road path planning via low-dimensional simulation | PPO, Curriculum Learning |
| Year | Published | Author | Title | Task | Github |
|---|---|---|---|---|---|
| 2020 | IV | PKU | Off-road Autonomous Vehicles Traversability Analysis and Trajectory Planning Based on Deep Inverse Reinforcement Learning | Reinforcement Learning, Traversability Estimation, Trajectory Planning | - |
| 2020 | ICRA | McGill University | Learning to Drive Off Road on Smooth Terrain in Unstructured Environments Using an On-Board Camera and Sparse Aerial Images | Reinforcement Learning, Traversability Estimation, Trajectory Planning | - |
| 2022 | RAL | NIIDT | AdaptiveON: Adaptive Outdoor Local Navigation Method for Stable and Reliable Actions | Reinforcement Learning, Sim-to-Real | https://github.com/jingGM/adaptiveON |
| 2024 | ICRA | Indiana University | Gaussian Process-based Traversability Analysis for Terrain Mapless Navigation | Sparse Gaussian Process (SGP) local map with a Rapidly-Exploring Random Tree* (RRT*) planner | https://github.com/abeleinin/gp-navigation |
If you find our survey useful in your research or applications, please consider giving us a star 🌟 and citing it in the following BibTeX entry.
@article{unstructuredAD,
title={Autonomous Ground Robots in Unstructured Environments: How Far Have We Come?},
author={Chen Min, Shubin Si, Xu Wang, Hanzhang Xue, Weizhong Jiang, Yang Liu, Juan Wang, Qingtian Zhu, Qi Zhu, Lun Luo, Fanjie Kong, Jinyu Miao, Xudong Cai, Shuai An, Wei Li, Jilin Mei, Tong Sun, Heng Zhai, Qifeng Liu, Fangzhou Zhao, Liang Chen, Shuai Wang, Erke Shang, Linzhi Shang, Kunlong Zhao, Fuyang Li, Hao Fu, Lei Jin, Jian Zhao, Fangyuan Mao, Zhipeng Xiao, Chengyang Li, Bin Dai, Dawei Zhao, Liang Xiao, Yiming Nie, Yu Hu},
journal={arXiv preprint arXiv:2410.07701},
year={2024}
}