Haptic-Aided Navigation Vehicle: Enhancing Obstacle Detection in Blind Spots and Transparent Object Scenarios

doi:10.1007/s12204-025-2807-7

Abstract

Abstract: As autonomous mobile robots are increasingly deployed in complex environments, traditional vision sensors and LiDAR encounter considerable limitations, particularly in detecting obstacles in blind spots or transparent objects. To address the issue of blind spots, we design a specialized haptic sensing structure and develop the haptic-aided navigation vehicle (HANV). This system integrates haptic sensors and LiDAR to deliver comprehensive perception, significantly enhancing close-range obstacle detection in areas that are typically beyond the range of conventional sensors. To tackle the challenge of transparent obstacles, which are often undetected by both vision and LiDAR sensors, we employ a fusion of haptic sensors and LiDAR. The haptic system provides physical contact feedback, ensuring reliable detection of transparent obstacles such as glass, while LiDAR offers long-range sensing capabilities. This combination enables HANV to navigate effectively in environments with transparent obstacles, overcoming the limitations of traditional sensing systems. Experiment results indicate that the proposed haptic and LiDAR integration substantially improves obstacle detection in both blind spots and environments with transparent obstacles. HANV achieves high success rates, minimal collisions, and efficient obstacle avoidance, particularly excelling in complex, confined spaces where conventional systems prove inadequate. These findings emphasize the efficacy of our approach in enhancing navigation performance in dynamic and challenging environments.

Key words: haptic sensor, LiDAR, autonomous obstacle avoidance, deep reinforcement learning, multimodal sensor fusion, mobile robot navigation

摘要： 随着自主移动机器人在复杂环境中的广泛应用，传统的视觉传感器和激光雷达（LiDAR）面临着显著的局限性，特别是在盲区和透明物体的障碍物检测方面。为了解决盲区问题，我们设计了一种专门的触觉传感结构，并开发了触觉辅助导航车辆（HANV）。该系统集成了触觉传感器和激光雷达，提供全面的感知能力，显著提高了在传统传感器无法覆盖的盲区区域的近距离障碍物检测能力。为了应对透明障碍物的挑战，这些障碍物通常无法被视觉和激光雷达传感器检测到，我们采用了触觉传感器与激光雷达的融合。触觉系统提供物理接触反馈，确保可靠地检测到玻璃等透明障碍物，而激光雷达则提供远程感知能力。两者的结合使得触觉辅助导航车辆能够有效地在具有透明障碍物的环境中导航，克服了传统传感系统的局限性。实验结果表明，所提出的触觉与激光雷达融合方案显著提高了在盲区和透明障碍物环境中的障碍物检测能力。触觉辅助导航车辆在复杂且狭小的空间中，特别是在传统系统表现不足的情况下，取得了较高的成功率、最小的碰撞率和高效的障碍物规避。这些结果突显了我们的方法在动态和具有挑战性的环境中提升导航性能的有效性。

关键词: 触觉传感器，激光雷达，自主避障，深度强化学习，多模态传感器融合，移动机器人导航

CLC Number:

TP242

Li Mingwang, Li Xinde, Zhang Zhentong, Wang Zeyu, Zhao Haoming. Haptic-Aided Navigation Vehicle: Enhancing Obstacle Detection in Blind Spots and Transparent Object Scenarios[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(1): 167-175.

References

[1] TONG J, XING X Y, GUO R Q, et al. Performance limitations analysis of visual sensors in low light conditions based on field test [C]// SAE 2022 Intelligent and Connected Vehicles Symposium. Shanghai: SAE, 2022: 2022-01-7086.

[2] ZHU K, ZHANG T. Deep reinforcement learning based mobile robot navigation: A review [J]. Tsinghua Science and Technology, 2021, 26(5): 674-691.

[3] CHEN Q P, XIE Y F, GUO S F, et al. Sensing system of environmental perception technologies for driverless vehicle: A review of state of the art and challenges [J]. Sensors and Actuators A: Physical, 2021, 319: 112566.

[4] PAWAR K S, TELI S N, SHETYE P, et al. Blind-spot monitoring system using lidar [J]. Journal of the Institution of Engineers (India): Series C, 2022, 103(5): 1071-1082.

[5] KWON D, MALAIYA R, YOON G, et al. A study on development of the camera-based blind spot detection system using the deep learning methodology [J]. Applied Sciences, 2019, 9(14): 2941.

[6] LUO C M, YANG S X, MO H W, et al. Safety aware robot coverage motion planning with virtual-obstacle-based navigation [C]//2015 IEEE International Conference on Information and Automation. Lijiang: IEEE, 2015: 2110-2115.

[7] LI X D, ZHANG X L, ZHU B, et al. A visual navigation method of mobile robot using a sketched semantic map [J]. International Journal of Advanced Robotic Systems, 2012, 9(4): 138.

[8] LUNECKAS M, LUNECKAS T, UDRIS D, et al. A hybrid tactile sensor-based obstacle overcoming method for hexapod walking robots [J]. Intelligent Service Robotics, 2021, 14(1): 9-24.

[9] LE N M D, NGUYEN N H, NGUYEN D A, et al. ViART: Vision-based soft tactile sensing for autonomous robotic vehicles [J]. IEEE/ASME Transactions on Mechatronics, 2024, 29(2): 1420-1430.

[10] KOSSAS T, REMMAS W, GKLIVA R, et al. Whisker-based tactile navigation algorithm for underground robots [C]//2024 IEEE International Conference on Robotics and Automation. Yokohama. IEEE, 2024: 13164-13170.

[11] YU Z Q, GUO Y, SU J J, et al. Bioinspired, multifunctional, active whisker sensors for tactile sensing of mobile robots [J]. IEEE Robotics and Automation Letters, 2022, 7(4): 9565-9572.

[12] OTA K, JHA D K, JATAVALLABHULA K M, et al. Tactile estimation of extrinsic contact patch for stable placement [C]//2024 IEEE International Conference on Robotics and Automation. Yokohama: IEEE, 2024: 13876-13882.

[13] LI Y, IBANEZ-GUZMAN J. Lidar for autonomous driving: The principles, challenges, and trends for automotive lidar and perception systems [J]. IEEE Signal Processing Magazine, 2020, 37(4): 50-61.

[14] LI X D, LUO C M, XU Y F, et al. A Fuzzy PID controller applied in AGV control system [C]//2016 International Conference on Advanced Robotics and Mechatronics. Macau: IEEE, 2016: 555-560.

[15] WANG Y, KANG J R, CHEN Z H, et al. Terrestrial locomotion of PogoX: From hardware design to energy shaping and step-to-step dynamics based control [C]//2024 IEEE International Conference on Robotics and Automation. Yokohama: IEEE, 2024: 3419-3425.

[16] LIU M, LI Z, YANG X D, et al. Dynamic analysis of a deployable/retractable damped cantilever beam [J]. Applied Mathematics and Mechanics, 2020, 41(9): 1321-1332.

[17] FUJIMOTO S, HOOF H V, MEGER D. Addressing function approximation error in actor-critic methods [C]// 35th International Conference on Machine Learning. Stockholm, PMLR, 2018: 1587-1596.

[18] CIMURS R, SUH I H, LEE J H. Goal-driven autonomous exploration through deep reinforcement learning [J]. IEEE Robotics and Automation Letters, 2022, 7(2): 730-737.

[19] LI Y J, WEI C Y, XIA Y. Lightweight multimodal fusion for autonomous navigation via deep reinforcement learning [M]// Proceedings of 3rd 2023 International Conference on Autonomous Unmanned Systems (3rd ICAUS 2023). Singapore: Springer, 2024: 78-87.

[20] LI X H, LI X D, KHYAM M O, et al. Visual navigation method for indoor mobile robot based on extended BoW model [J]. CAAI Transactions on Intelligence Technology, 2017, 2(4): 142-147.