ABSTRACT
In order to solve the problem of terrain recognition in the case of sample data and improve the terrain detection ability of unmanned platform in complex environment, a visual touch fusion idea is put forward. This idea is based on the original broad learning, by establishing a multimodal cascade feature node broad learning framework for terrain detection. First, we extract the initial features of touch and vision, and fuse them to obtain a fusion feature matrix. Then, the broad learning classifier is used to process the fusion feature matrix and get the result of terrain recognition. Finally, we use a self-built visual-tactile terrain dataset for experimental verification. The experimental results show that CFBRL-KCCA algorithm proposed gets the highest classification among several algorithms. As a result, it reached 84.67 percent. This provides an effective strategy for terrain detection by unmanned platforms in complex environments.
- Jiancheng Fang, Xiaolin Ning, Xin Ma, Jin Liu, and Mingzhen Gui. 2018. A survey of autonomous astronomical navigation technology for deep space detectors. Flight Control & Detection, 1(1), 1-15.Google Scholar
- Caixia Liu, Jianjun Fang, Yanxia Liu, and Yujiao Lu. 2017. The field terrain recognition based on extreme learning machine using wavelet features. In 2017 IEEE International Conference on Mechatronics and Automation (ICMA). IEEE, 1947-1951.Google ScholarDigital Library
- Baoyin Hexi and Pengbin Ma. 2018. Overview of autonomous navigation method for mars probe. Flight control & detection, 1(1), 34-40.Google Scholar
- Dandan Dong. 2016. Research and implementation of terrain classification method based on vision and force signal[D]. Southeast University.Google Scholar
- Zhengyuan Hu and Weijun Tao. 2023. Research on local obstacle avoidance method in complex terrain and obstacle environment. modern electrics techniques 46, 15 (2023), 180-186.Google Scholar
- Quanfeng Li, Haibo Wu, Jiang Chen, and Yixiao Zhang. 2023. Gmapping Mapping Based on Lidar and RGB-D Camera Fusion[J]. LASER & OPTOELECTRONICS PROGRESS 60, 12 (2023). DOI:10.3788/LOP221447.Google ScholarCross Ref
- Yutong Xie, Hao Chen, Wanzhi Zhang, Shuaijun Li, and Huihuan Qian. 2018. Lidar-IMU fusion for 2.5D mapping. In Proceedings of the 2018 IEEE International Conference on Robotics and Biomimetics (ROBIO). IEEE, 832-837.Google ScholarDigital Library
- Chaoqun Wang, Jiankun Wang, Chenming Li, Danny Ho, Jiyu Cheng, Tiekun Yan, Lili Meng, and Max Q.-h. Meng. 2019. Safe and robust mobile robot navigation in uneven indoor environments. Sensors 19, 13 (July 2019), 2993. DOI:https://doi.org/10.3390/s19132993.Google ScholarCross Ref
- Paul Filitchkin and Katie Byl. 2012. Feature-based terrain classification for littledog. In Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 1387-1392Google ScholarCross Ref
- Ingmar Posner, Mark Cummins, and Paul Newman. 2009. A generative framework for fast urban labeling using spatial and temporal context. Autonomous Robots 26, 2–3 (March 2009), 153–170. DOI:https://doi.org/10.1007/s10514-009-9110-6.Google ScholarDigital Library
- Csaba Kertész. 2016. Rigidity-Based surface recognition for a domestic legged robot. IEEE Robotics and Automation Letters 1, 1 (January 2016), 309–315. DOI:https://doi.org/10.1109/lra.2016.2519949.Google ScholarCross Ref
- Jakub Mrva and Jan Faigl. 2015. Feature Extraction for Terrain Classification with Crawling Robots. ITAT (January 2015), 179–185. Retrieved from http://ceur-ws.org/Vol-1422/179.pdf.Google Scholar
- Matej Hoffmann, Karla Štěpánová, and R. Michal. 2014. The effect of motor action and different sensory modalities on terrain classification in a quadruped robot running with multiple gaits. Robotics and Autonomous Systems 62, 12 (December 2014), 1790–1798. DOI:https://doi.org/10.1016/j.robot.2014.07.006.Google ScholarDigital Library
- Meichao Xing. 2023. Research on terrain recognition algorithm based on hexapod robot foot sensor[D]. Jilin University.Google Scholar
- Caixia Liu, Jianjun Fang, Yanxia Liu, and Yanwu Lü. 2018. Field terrain recognition based on extreme learning theory using wavelet and texture features. Advances in Mechanical Engineering (May 2018). DOI:https://doi.org/10.1177/1687814018772713.Google ScholarCross Ref
- C. L. Philip Chen and Zhulin Liu. 2018. Broad Learning system: An effective and efficient incremental learning system without the need for deep architecture. IEEE Transactions on Neural Networks and Learning Systems 29, 1 (January 2018), 10–24. DOI:https://doi.org/10.1109/tnnls.2017.2716952.Google ScholarCross Ref
- C. L. Philip Chen, Zhulin Liu, and Feng Shuang. 2019. Universal approximation capability of broad learning system and its structural variations. IEEE Transactions on Neural Networks and Learning Systems 30, 4 (April 2019), 1191–1204. DOI:https://doi.org/10.1109/tnnls.2018.2866622.Google ScholarCross Ref
- Wen Wang, Huaping Liu, Lianzhi Yu, and Fuchun Sun. 2014. Human activity recognition using smart phone embedded sensors: A linear dynamical systems method. In Proceedings of the 2014 International Joint Conference on Neural Networks (IJCNN). IEEE, 1185-1190Google ScholarCross Ref
- Guang-Bin Huang, Qin-Yu Zhu, and Chee Kheong Siew. 2006. Extreme learning machine: Theory and applications. Neurocomputing 70, 1–3 (December 2006), 489–501. DOI:https://doi.org/10.1016/j.neucom.2005.12.126.Google ScholarCross Ref
Index Terms
- Multimodal Fusion Method for Complex Terrain Detection of Unmanned Platform
Recommendations
3D object detection based on the fusion of projected point cloud and image features
EITCE '22: Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer EngineeringThe complementary advantages of point cloud and image can provide more accurate 3D and semantic information to the model. Aiming at the problems that most existing methods adopt a single fusion strategy and thus fail to achieve deep fusion of image and ...
Research on Feature Extraction and Multimodal Fusion of Video Caption Based on Deep Learning
ICMSS 2020: Proceedings of the 2020 4th International Conference on Management Engineering, Software Engineering and Service SciencesVideo Caption shows the objects, attributes and their relationship in natural language. It has been a very challenging research topic in the field of computer and multimedia. In this paper, the method of deep learning is used to extract the video frame ...
Multimodal fusion: a new hybrid strategy for dialogue systems
ICMI '06: Proceedings of the 8th international conference on Multimodal interfacesThis is a new hybrid fusion strategy based primarily on the implementation of two former and differentiated approaches to multimodal fusion [11] in multimodal dialogue systems. Both approaches, their predecessors and their respective advantages and ...
Comments