4.7 Review

A comprehensive survey of unmanned ground vehicle terrain traversability for unstructured environments and sensor technology insights

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Computer Science, Artificial Intelligence

On the use of information fusion techniques to improve information quality: Taxonomy, opportunities and challenges

Raul Gutierrez et al.

Summary: Information fusion is a promising method that combines heterogeneous information sources to improve the quality of information processing. However, challenges remain in understanding the impact of different fusion methods on information quality, evaluating information quality, and ensuring flexibility in the fusion process to adapt to dynamic environments.

INFORMATION FUSION (2022)

Article Robotics

On the role of feature and signal selection for terrain learning in planetary exploration robots

Angelo Ugenti et al.

Summary: This study presents a novel approach to terrain classification for planetary exploration rovers using proprioceptive sensing, achieving high accuracy in predicting terrain types. By comparing support vector machine and deep convolutional neural network learning methods, it is shown that proprioceptive sensing is effective and CNN outperforms SVM in extrapolation tasks.

JOURNAL OF FIELD ROBOTICS (2022)

Proceedings Paper Computer Science, Artificial Intelligence

Predicting Autonomous Vehicle Navigation Parameters via Image and Image-and-Point Cloud Fusion-based End-to-End Methods

Semih Beycimen et al.

Summary: This paper presents a study on end-to-end methods for predicting autonomous vehicle navigation parameters. Image-based and Image & Lidar points-based models were trained and tested under Nvidia architectures as well as Densenet-169, Resnet-152, and Inception-v4. The study investigated various learning parameters, input models, and pre-processing algorithms, and selected the best ones for the main framework. Results reveal that the Nvidia architecture trained Image & Lidar points-based method offers better accuracy rate for steering angle and speed.

2022 IEEE INTERNATIONAL CONFERENCE ON MULTISENSOR FUSION AND INTEGRATION FOR INTELLIGENT SYSTEMS (MFI) (2022)

Article Automation & Control Systems

Learning from experience for rapid generation of local car maneuvers

Piotr Kicki et al.

Summary: For autonomous driving, rapidly responding to changing scenes and generating feasible paths is crucial. Researchers have proposed a method using deep neural networks for path planning, which not only completes quickly but also performs well in real and simulated scenarios, generating smooth paths.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2021)

Article Engineering, Environmental

Recurrent and convolutional neural networks for deep terrain classification by autonomous robots

Fabio Vulpi et al.

Summary: The main challenge for field robots in the future is to increase autonomy for long distance and duration applications. One hypothesis is that terrain can be effectively detected using proprioceptive signals without relying on visual or depth information. Another hypothesis suggests that artificial deep neural networks can provide accurate and robust solution for classifying different terrain types.

JOURNAL OF TERRAMECHANICS (2021)

Review Chemistry, Analytical

Learning-Based Methods of Perception and Navigation for Ground Vehicles in Unstructured Environments: A Review

Dario Calogero Guastella et al.

Summary: The review focuses on recent contributions in robotics research that use learning-based methods to address the issue of autonomous perception and interpretation for ground vehicles in unstructured environments. Perception is highlighted as crucial for autonomous navigation, providing necessary information for accurate navigation in complex environments. This review is the first of its kind in this context, shedding light on the importance of perception for context-aware navigation.

SENSORS (2021)

Article Chemistry, Analytical

Weakly-Supervised Recommended Traversable Area Segmentation Using Automatically Labeled Images for Autonomous Driving in Pedestrian Environment with No Edges

Yuya Onozuka et al.

Summary: In this paper, a method for weakly-supervised recommended traversable area segmentation in environments with no edges using automatically labeled images based on paths selected by humans is proposed. Evaluation results showed the effectiveness of the proposed learning methods for recommended traversable area detection in edge-less environments. This approach is useful for detecting recommended areas in environments where roads and sidewalks coexist.

SENSORS (2021)

Article Computer Science, Information Systems

On-Line Learning and Updating Unmanned Tracked Vehicle Dynamics

Natalia Strawa et al.

Summary: This paper presents a comprehensive mathematical model of a skid steering tracked vehicle and designs a control law to address performance requirements for unmanned ground vehicles (UGV). Analysis of the controller under model uncertainties reveals undesirable behavior, and a compound identification scheme is proposed to estimate model parameters and improve adaptive capability for the control system.

ELECTRONICS (2021)

Article Computer Science, Artificial Intelligence

Three-Dimensional Mapping with Augmented Navigation Cost through Deep Learning

Felipe G. Oliveira et al.

Summary: This study focuses on the mapping of terrain features based on inertial and LiDAR measurements to estimate navigation cost for autonomous ground robots. The proposed methodology includes five main steps and experiments conducted validate its effectiveness and reliability in different outdoor environments with varying terrain characteristics.

JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS (2021)

Article Automation & Control Systems

Autonomous vehicle path tracking control considering the stability under lane change

Xiaolong Chen et al.

Summary: This article presents an approach to improve path tracking and stabilization on low friction coefficient roads at high speeds, including estimating vehicle and road states using the unscented Kalman filter, implementing predictive control for path tracking, and designing a steering angle compensation controller.

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART I-JOURNAL OF SYSTEMS AND CONTROL ENGINEERING (2021)

Review Chemistry, Analytical

Sensor and Sensor Fusion Technology in Autonomous Vehicles: A Review

De Jong Yeong et al.

Summary: Automated driving is becoming a pivotal technology in transforming the future of transportation with advancements in sensor and communication technology. Sensor calibration is crucial for ensuring the safety and feasibility of automated driving vehicles, while sensor fusion and obstacle detection techniques play a critical role in the successful implementation of autonomous driving systems.

SENSORS (2021)

Article Robotics

BADGR: An Autonomous Self-Supervised Learning-Based Navigation System

Gregory Kahn et al.

Summary: Traditional mobile robot navigation solutions focus on the geometric structure of the environment, but this approach may not always be effective. BADGR utilizes a reinforcement learning approach to move beyond purely geometric navigation solutions, learning physical navigational affordances in order to navigate mobile robots without the need for simulation or human supervision.

IEEE ROBOTICS AND AUTOMATION LETTERS (2021)

Article Robotics

LiDAR-based robust localization for field autonomous vehicles in off-road environments

Ruike Ren et al.

Summary: This article introduces a reliable degeneracy indicator for evaluating scan matching performance in off-road environments. The indicator is integrated into a factor graph optimization framework, improving the system's ability to handle incomplete and outdated LiDAR maps. Extensive testing on real-world data sets shows that the proposed system outperforms existing methods in degenerate scenarios.

JOURNAL OF FIELD ROBOTICS (2021)

Article Chemistry, Analytical

An Overview of Autonomous Vehicles Sensors and Their Vulnerability to Weather Conditions

Jorge Vargas et al.

Summary: This paper discusses the impacts of different weather conditions on the perception systems of autonomous vehicles, analyzing the principles and performance adjustments of various common AV sensors. By quantifying these impacts, more realistic scenarios can be simulated and more accurate object identification models can be constructed in the physical world to assist AVs in effectively perceiving their surrounding environment.

SENSORS (2021)

Article Agronomy

Evaluation of Depth Cameras for Use in Fruit Localization and Sizing: Finding a Successor to Kinect v2

Chiranjivi Neupane et al.

Summary: For applications involving fruit localization and sizing in orchards, ToF cameras such as Blaze 101 or Azure Kinect are recommended for their ability to meet the 20 mm root mean square error standard in sunlight and orchard conditions. It is suggested that future assessments of depth cameras for this application should include bias-corrected RMSE estimation and assessment of bias on estimated camera-to-fruit distances at various intervals under different lighting conditions.

AGRONOMY-BASEL (2021)

Article Engineering, Civil

Memory-Based Deep Reinforcement Learning for Obstacle Avoidance in UAV With Limited Environment Knowledge

Abhik Singla et al.

Summary: This paper presents a method for enabling a UAV quadrotor to autonomously avoid collisions with obstacles in unstructured and unknown indoor environments. The method, based on deep reinforcement learning, uses recurrent neural networks with temporal attention and outperforms prior works in terms of distance covered without collisions. Additionally, the technique reduces power wastage by minimizing oscillatory motion of the UAV.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2021)

Proceedings Paper Automation & Control Systems

Traversability Mapping in Off-Road Environment using Semantic Segmentation

Lalitha Dabbiru et al.

Summary: Developing autonomous navigation algorithms in off-road environments requires labeled data for training and testing machine learning algorithms. This research focuses on collecting and annotating 2D camera data in off-road environments, and applying semantic segmentation algorithms to predict trajectory based on the type of ground vehicle. The ground vehicles considered include sedan, pickup truck, and military vehicles, with data collected at the CAVS proving grounds during different seasons.

AUTONOMOUS SYSTEMS: SENSORS, PROCESSING, AND SECURITY FOR VEHICLES AND INFRASTRUCTURE 2021 (2021)

Article Engineering, Electrical & Electronic

GEM: Online Globally Consistent Dense Elevation Mapping for Unstructured Terrain

Yiyuan Pan et al.

Summary: The GEM system proposed in this article generates dense local and global elevation maps, maintains global consistency by updating relative poses between submaps, accelerates local mapping by integrating traversability analysis, and ensures real-time performance through CPU-GPU coordinated processing. Substantial experimental results validate the efficiency and effectiveness of GEM.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2021)

Article Computer Science, Artificial Intelligence

Self-Supervised Model Adaptation for Multimodal Semantic Segmentation

Abhinav Valada et al.

INTERNATIONAL JOURNAL OF COMPUTER VISION (2020)

Article Robotics

Terrain-Aware Path Planning and Map Update for Mars Sample Return Mission

Gabrielle Hedrick et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

Article Robotics

Deep Reinforcement Learning for Safe Local Planning of a Ground Vehicle in Unknown Rough Terrain

Shirel Josef et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

Proceedings Paper Automation & Control Systems

Learning to Drive Off Road on Smooth Terrain in Unstructured Environments Using an On-Board Camera and Sparse Aerial Images

Travis Manderson et al.

2020 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA) (2020)

Proceedings Paper Automation & Control Systems

RGB-D Based Semantic SLAM Framework for Rescue Robot

Wenbang Deng et al.

2020 CHINESE AUTOMATION CONGRESS (CAC 2020) (2020)

Proceedings Paper Automation & Control Systems

Deep Reinforcement Learning-based ROS-Controlled RC Car for Autonomous Path Exploration in the Unknown Environment

Sabir Hossain et al.

2020 20TH INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND SYSTEMS (ICCAS) (2020)

Proceedings Paper Automation & Control Systems

SVM based sensor fusion for improved terrain classification

Akhil Kurup et al.

AUTONOMOUS SYSTEMS: SENSORS, PROCESSING, AND SECURITY FOR VEHICLES AND INFRASTRUCTURE 2020 (2020)

Proceedings Paper Computer Science, Theory & Methods

Deep Reinforcement Learning For Visual Navigation of Wheeled Mobile Robots

Ezebuugo Nwaonumah et al.

IEEE SOUTHEASTCON 2020 (2020)

Proceedings Paper Automation & Control Systems

Assuring the Safety of End-to-End Learning-Based Autonomous Driving through Runtime Monitoring

Joerg Grieser et al.

2020 23RD EUROMICRO CONFERENCE ON DIGITAL SYSTEM DESIGN (DSD 2020) (2020)

Article Robotics

Gated Recurrent Fusion to Learn Driving Behavior from Temporal Multimodal Data

Athma Narayanan et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2020)

Article Computer Science, Information Systems

Multi-Sensor Fusion in Automated Driving: A Survey

Zhangjing Wang et al.

IEEE ACCESS (2020)

Article Automation & Control Systems

AutoRally AN OPEN PLATFORM FOR AGGRESSIVE AUTONOMOUS DRIVING

Brian Goldfain et al.

IEEE CONTROL SYSTEMS MAGAZINE (2019)

Article Engineering, Civil

Vision-Based Navigation of Autonomous Vehicles in Roadway Environments with Unexpected Hazards

Mhafuzul Islam et al.

TRANSPORTATION RESEARCH RECORD (2019)

Article Automation & Control Systems

LIDAR-camera fusion for road detection using fully convolutional neural networks

Luca Caltagirone et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2019)

Proceedings Paper Automation & Control Systems

Improving Traversability Estimation Through Autonomous Robot Experimentation

Christos Sevastopoulos et al.

COMPUTER VISION SYSTEMS (ICVS 2019) (2019)

Proceedings Paper Automation & Control Systems

Optimization-Based Terrain Analysis and Path Planning in Unstructured Environments

Ueli Graf et al.

2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA) (2019)

Proceedings Paper Automation & Control Systems

Variational End-to-End Navigation and Localization

Alexander Amini et al.

2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA) (2019)

Article Robotics

Where Should I Walk? Predicting Terrain Properties From Images Via Self-Supervised Learning

Lorenz Wellhausen et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2019)

Article Automation & Control Systems

Machine-learning based approaches for self-tuning trajectory tracking controllers under terrain changes in repetitive tasks

Alvaro Javier Prado et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2018)

Article Engineering, Civil

ERFNet: Efficient Residual Factorized ConvNet for Real-Time Semantic Segmentation

Eduardo Romera et al.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2018)

Article Engineering, Civil

Learning Traversability From Point Clouds in Challenging Scenarios

Mauro Bellone et al.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2018)

Article Engineering, Civil

Scene-Adaptive Off-Road Detection Using a Monocular Camera

Jilin Mei et al.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2018)

Article Computer Science, Artificial Intelligence

Direct Sparse Odometry

Jakob Engel et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2018)

Article Engineering, Environmental

Terrain classification using ToF sensors for the enhancement of agricultural machinery traversability

Francisco Yandun Narvaez et al.

JOURNAL OF TERRAMECHANICS (2018)

Article Computer Science, Artificial Intelligence

3-D LiDAR + Monocular Camera: An Inverse-Depth-Induced Fusion Framework for Urban Road Detection

Shuo Gu et al.

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES (2018)

Proceedings Paper Computer Science, Artificial Intelligence

LiDAR-Video Driving Dataset: Learning Driving Policies Effectively

Yiping Chen et al.

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) (2018)

Article Robotics

1 year, 1000 km: The Oxford RobotCar dataset

Will Maddern et al.

INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH (2017)

Article Robotics

Large-scale cost function learning for path planning using deep inverse reinforcement learning

Markus Wulfmeier et al.

INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH (2017)

Proceedings Paper Computer Science, Artificial Intelligence

Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention

Jinkyu Kim et al.

2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV) (2017)

Article Agricultural Engineering

Terrain assessment for precision agriculture using vehicle dynamic modelling

Giulio Reina et al.

BIOSYSTEMS ENGINEERING (2017)

Proceedings Paper Computer Science, Artificial Intelligence

ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes

Angela Dai et al.

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) (2017)

Article Engineering, Mechanical

Slip-based terrain estimation with a skid-steer vehicle

Giulio Reina et al.

VEHICLE SYSTEM DYNAMICS (2016)

Article Computer Science, Artificial Intelligence

A Kalman-filtering-based Approach for Improving Terrain Mapping in off-road Autonomous Vehicles

Isao Parra-Tsunekawa et al.

JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS (2015)

Article Automation & Control Systems

Visual ground segmentation by radar supervision

Annalisa Milella et al.

ROBOTICS AND AUTONOMOUS SYSTEMS (2014)

Article Automation & Control Systems

Terrain traversability analysis methods for unmanned ground vehicles: A survey

Panagiotis Papadakis

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2013)

Article Robotics

Terrain Adaptive Navigation for Planetary Rovers

Daniel Helmick et al.

JOURNAL OF FIELD ROBOTICS (2009)

Article Robotics

Learning and prediction of slip from visual information

Anelia Angelova et al.

JOURNAL OF FIELD ROBOTICS (2007)

Article Robotics

Natural terrain classification using three-dimensional ladar data for ground robot mobility

Jean-Francois Lalonde et al.

JOURNAL OF FIELD ROBOTICS (2006)

Article Robotics

Vision-based terrain characterization and traversability assessment

A Howard et al.

JOURNAL OF ROBOTIC SYSTEMS (2001)