Publication | Yinchuan Wang

2025

TCDS
SELM: From Efficient Autonomous Exploration to Long-term Monitoring in Semantic Level

Fang Lang, Yongsen Qin, Yinchuan Wang, and 5 more authors

IEEE Transactions on Cognitive and Developmental Systems, Jan 2025

Abs Bib HTML

Maintaining up-to-date environmental models from initial deployment through long-term autonomy in service is critical for applications like navigation and task planning. To address the challenges of persistent monitoring in unknown environments, we introduce a Two-Stage monitoring strategy, termed the Semantic-level autonomous Exploration and Long-term environment Monitoring (SELM) framework. In the first stage, we introduce a novel semantic exploration method to adapt to new environments quickly. Leveraging the semantic information within the incrementally constructed 3D Scene Graph (3DSG), we combine the Next-Best-View (NBV) selection with room semantics, introducing a more efficient and comprehensive approach for multi-room indoor environment exploration. In addition, the exploration provides patrol routes, the room distance-connectivity graph, and complete environment initial states for subsequential monitoring. The monitoring stage aims to persistently patrol to update the world model in the presence of dynamic changes, including changes in objects’ positions. We formulate the long-term monitoring problem as the Partially Observable Markov Decision Process (POMDP) to cope with the environmental uncertainty. To solve the POMDP, we propose the Graph Attention Bidirectional long short-term memory Proximal Policy Optimization (GABPPO) algorithm for the optimal patrol strategy. The feasibility and effectiveness of the proposed SELM framework are verified through extensive experiments.
@article{lang2025selm, title = {SELM: From Efficient Autonomous Exploration to Long-term Monitoring in Semantic Level}, author = {Lang, Fang and Qin, Yongsen and Wang, Yinchuan and Liu, Jin and Wang, Chaoqun and Song, Wei and Zhu, Qiuguo and Song, Rui}, journal = {IEEE Transactions on Cognitive and Developmental Systems}, year = {2025}, month = jan, doi = {10.1109/TCDS.2025.3531367}, dimensions = {false}, google_scholar_id = {eQOLeE2rZwMC}, }
J-FR
ROLO-SLAM: Rotation-Optimized LiDAR-Only SLAM in Uneven Terrain with Ground Vehicle

Yinchuan Wang, Bin Ren, Xiang Zhang, and 5 more authors

Journal of Field Robotics, Jan 2025

Abs arXiv Bib HTML Video Code

LiDAR-based SLAM is recognized as one effective method to offer localization guidance in rough environments. However, off-the-shelf LiDAR-based SLAM methods suffer from significant pose estimation drifts, particularly components relevant to the vertical direction, when passing to uneven terrains. This deficiency typically leads to a conspicuously distorted global map. In this article, a LiDAR-based SLAM method is presented to improve the accuracy of pose estimations for ground vehicles in rough terrains, which is termed Rotation-Optimized LiDAR-Only (ROLO) SLAM. The method exploits a forward location prediction to coarsely eliminate the location difference of consecutive scans, thereby enabling separate and accurate determination of the location and orientation at the front-end. Furthermore, we adopt a parallel-capable spatial voxelization for correspondence-matching. We develop a spherical alignment-guided rotation registration within each voxel to estimate the rotation of vehicle. By incorporating geometric alignment, we introduce the motion constraint into the optimization formulation to enhance the rapid and effective estimation of LiDAR’s translation. Subsequently, we extract several keyframes to construct the submap and exploit an alignment from the current scan to the submap for precise pose estimation. Meanwhile, a global-scale factor graph is established to aid in the reduction of cumulative errors. In various scenes, diverse experiments have been conducted to evaluate our method. The results demonstrate that ROLO-SLAM excels in pose estimation of ground vehicles and outperforms existing state-of-the-art LiDAR SLAM frameworks.
@article{wang2025rolo, title = {ROLO-SLAM: Rotation-Optimized LiDAR-Only SLAM in Uneven Terrain with Ground Vehicle}, author = {Wang, Yinchuan and Ren, Bin and Zhang, Xiang and Wang, Pengyu and Wang, Chaoqun and Song, Rui and Li, Yibin and Q.-H., Meng Max}, journal = {Journal of Field Robotics}, year = {2025}, month = jan, doi = {10.1002/rob.22505}, dimensions = {false}, google_scholar_id = {YsMSGLbcyi4C}, }

2024

IEEE T-IV
Transformer-based Traversability Analysis for Autonomous Navigation in Outdoor Environments with Water Hazard

Mingrui Xu, Yinchuan Wang, Xiang Zhang, and 4 more authors

IEEE Transactions on Intelligent Vehicles, Jun 2024

Abs Bib HTML Video

Traversability analysis is of vital importance for autonomous vehicle navigation in outdoor environments. This paper focuses on the navigation algorithms in outdoor environments with water hazards, where the water areas can neither simply be neglected nor be deemed as non-traversable for safe and efficient navigation. To this end, we present a tightly-integrated navigation algorithms framework, which includes an environmental traversability analysis model and a control signal acquisition model. The traversability analysis model includes the SwinT-CGRU semantic segmentation model and the ConvCRFs water analysis model proposed in this paper. SwinT-CGRU can efficiently extract the feature from the input successive images, providing initial analysis of environmental traversability. To further analyze the water region, we employ a ConvCRFs, which determines the water traversability in an effective way. The results of the water traversability analysis are convincing for reliable vehicle navigation in outdoor scenarios. Based on the traversability analysis results, we propose a path planning approach by adapting the artificial potential field method to the image space. The control signal for steering and driving the vehicle can be acquired through an established mapping from the image space to the vehicle space. Experiments demonstrate that our segmentation model improves mIoU by 2.09-18.21% on the RELLIS-3D dataset. Furthermore, our navigation algorithm exhibits real-time performance and stability, ensuring the safety of vehicles in outdoor environments.
@article{xu2024transformer, title = {Transformer-based Traversability Analysis for Autonomous Navigation in Outdoor Environments with Water Hazard}, author = {Xu, Mingrui and Wang, Yinchuan and Zhang, Xiang and Mao, Donghui and Wang, Chaoqun and Song, Rui and Li, Yibin}, journal = {IEEE Transactions on Intelligent Vehicles}, year = {2024}, month = jun, doi = {10.1109/TIV.2024.3419846}, dimensions = {false}, google_scholar_id = {Tyk-4Ss8FVUC}, }
ICRA2024
History-Aware Planning for Risk-free Autonomous Navigation on Unknown Uneven Terrain

Yinchuan Wang, Nianfei Du, Yongsen Qin, and 3 more authors

2024 IEEE International Conference on Robotics and Automation (ICRA 2024), Japan Yokohama, May 2024

Abs arXiv Bib HTML Video Code

It is challenging for the mobile robot to achieve autonomous and mapless navigation in the unknown environment with uneven terrain. In this study, we present a layered and systematic pipeline. At the local level, we maintain a tree structure that is dynamically extended with the navigation. This structure unifies the planning with the terrain identification. Besides, it contributes to explicitly identifying the hazardous areas on uneven terrain. In particular, certain nodes of the tree are consistently kept to form a sparse graph at the global level, which records the history of the exploration. A series of subgoals that can be obtained in the tree and the graph are utilized for leading the navigation. To determine a subgoal, we develop an evaluation method whose input elements can be efficiently obtained on the layered structure. We conduct both simulation and real-world experiments to evaluate the developed method and its key modules. The experimental results demonstrate the effectiveness and efficiency of our method. The robot can travel through the unknown uneven region safely and reach the target rapidly without a preconstructed map.
@article{wang2024history, title = {History-Aware Planning for Risk-free Autonomous Navigation on Unknown Uneven Terrain}, author = {Wang, Yinchuan and Du, Nianfei and Qin, Yongsen and Zhang, Xiang and Song, Rui and Wang, Chaoqun}, journal = {2024 IEEE International Conference on Robotics and Automation (ICRA 2024)}, location = {Japan Yokohama}, pages = {777--780}, year = {2024}, month = may, doi = {10.1109/ICRA57147.2024.10610488}, dimensions = {false}, google_scholar_id = {zYLM7Y9cAGgC}, }

2023

CCC2023

A Comfortable Interaction Strategy for the Visually Impaired Using Quadruped Guidance Robot

Nianfei Du, Yinchuan Wang, Zhengguo Zhu, and 3 more authors

2023 42nd Chinese Control Conference (CCC), China Tianjin, Jul 2023

Abs

A guidance robot that can guide its user to the destination smoothly and comfortably without rigorous training is expected to bring more convenience to the visually impaired. Although guidance robots have evolved considerably, little attention has been paid to the importance of human-robot interaction in the guidance process which will bring an unpleasant experience for the user when the amplitude of the interaction force increases. In this paper, we propose an interaction strategy for the visually impaired using quadruped guidance robot based on compliance control to make the user more comfortable to be guided. The user can track robot’s trajectory by perceiving the variation of traction force or manipulate the movement of the robot by applying force to the robot using proposed strategy. We also build a quadruped guidance robot system with an autonomous navigation algorithm. The robot can plan a safe path that ensures the user does not collide with obstacles when following the robot. Then the robot continuously interacts with its user while following the planned path to the goal to provide a comfortable experience. Experiments show that the interaction strategy we deployed in our quadruped guidance robot reduces traction force during guidance and improves the user experience.
BIROB

Autonomous battery-changing system for UAV’s lifelong flight

Jiyang Chen, Wenxi Li, Yingting Sha, and 5 more authors

Biomimetic Intelligence and Robotics, Jun 2023

Abs

Unmanned aerial vehicles (UAVs) lifelong flight is essential to accomplish various tasks, e.g., aerial patrol, aerial rescue, etc. However, traditional UAVs have limited power to sustain their flight and need skilled operators manually control their charging process. Manufacturers and users are eagerly seeking a reliable autonomous battery-changing solution. To address this need, we propose and design an autonomous battery-changing system for UAVs using the theory of inventive problem solving (TRIZ) and user-centered design (UCD) methods. For practical application, we employ UCD to thoroughly analyze user requirements, identify multiple pairs of technical contradictions, and solve these inconsistencies using TRIZ theory. Furthermore, we design an autonomous battery-changing hardware system that meets user requirements and realizes the battery change process after the automatic landing of UAVs. Finally, we conduct experiments to validate our system’s effectiveness. The experimental results show that our battery-changing system can implement the autonomous battery-changing task, and our system has high efficiency and robustness in the real environment.

2022

2022IROS
Low-drift LiDAR-only Odometry and Mapping for UGVs in Environments with Non-level Roads

Xiangyu Chen, Yinchuan Wang, Chaoqun Wang, and 2 more authors

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Japan Kyoto, Oct 2022

Abs Bib HTML Video

This study focuses on localization and mapping for UGVs when they are deployed in environments with non-level roads. In these scenarios, the vehicles need to travel through flat but not necessarily level grounds, i.e., ascent or descent, which may cause drifts of the robot pose and distortion of the map. We develop a low-drift LiDAR odometry and mapping approach for the UGV with LiDAR as the only exteroceptive sensor. A factor-graph based pose optimization method is developed with a specifically designed factor named slope factor. This factor includes the slope information that is estimated from a real-time LiDAR data stream. The slope information is also used to enhance the loop-closure detection procedure. Moreover, an incremental pitch estimation mechanism is designed to achieve further pose estimation refinement. We demonstrate the effectiveness of the developed framework in real-world environments. The odometry drift is lower and the map is more precise than experiments with the state-of-the-arts. Notably, on the Kitti dataset, our method also exhibits convincing performance, demonstrating its strength in more general application scenarios.
@article{chen2022low, title = {Low-drift LiDAR-only Odometry and Mapping for UGVs in Environments with Non-level Roads}, author = {Chen, Xiangyu and Wang, Yinchuan and Wang, Chaoqun and Song, Rui and Li, Yibin}, journal = {2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, location = {Japan Kyoto}, pages = {13174--13180}, year = {2022}, organization = {IEEE}, month = oct, doi = {10.1109/IROS47612.2022.9982264}, dimensions = {false}, google_scholar_id = {u5HHmVD_uO8C}, }
ROBIO2022
Information Map Prediction based on Learning Network for Reinforced Autonomous Exploration

Yinchuan Wang, Mingrui Xu, Xiangyu Chen, and 3 more authors

2022 IEEE International Conference on Robotics and Biomimetics (ROBIO), China Xishuangbanna, Dec 2022

Abs Bib HTML Video Code

Autonomous exploration is the prerequisite for reliable robot navigation in unknown environments via providing the necessary information. The goal is to build a precise environment map while at the same time locating the robot robustly. To achieve this goal, this paper develops an information map that jointly provides the information of mapping and localization uncertainty. It consists of a localization uncertainty map and a model uncertainty map. The former evaluates the landmarks around the robot and the latter indicates the map completeness based on information theory. The information map serves as the input for making better decisions of where-to-go to achieve higher localization and mapping performance. To improve the efficiency of information map prediction, we develop a learning-based map prediction framework, where a dataset is established autonomously. The effectiveness and efficiency of the developed framework are verified through experiments in simulation environments.
@article{wang2022information, title = {Information Map Prediction based on Learning Network for Reinforced Autonomous Exploration}, author = {Wang, Yinchuan and Xu, Mingrui and Chen, Xiangyu and Zhang, Xiang and Wang, Chaoqun and Song, Rui}, journal = {2022 IEEE International Conference on Robotics and Biomimetics (ROBIO)}, location = {China Xishuangbanna}, pages = {1982--1988}, year = {2022}, organization = {IEEE}, month = dec, doi = {10.1109/ROBIO55434.2022.10011800}, dimensions = {false}, google_scholar_id = {u-x6o8ySG0sC}, }