Articles | Volume 15, issue 1
https://doi.org/10.5194/ms-15-1-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/ms-15-1-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
03 Jan 2024
Research article |
| 03 Jan 2024
| 03 Jan 2024
Visual simultaneous localization and mapping (vSLAM) algorithm based on improved Vision Transformer semantic segmentation in dynamic scenes
Mengyuan Chen
CORRESPONDING AUTHOR
School of Electrical Engineering, Anhui Polytechnic University, Wuhu, China
Key Laboratory of Advanced Perception and Intelligent Control of High-end Equipment, Ministry of Education, Anhui, China
Hangrong Guo
School of Electrical Engineering, Anhui Polytechnic University, Wuhu, China
Runbang Qian
School of Electrical Engineering, Anhui Polytechnic University, Wuhu, China
Guangqiang Gong
School of Electrical Engineering, Anhui Polytechnic University, Wuhu, China
Hao Cheng
School of Electrical Engineering, Anhui Polytechnic University, Wuhu, China
Related authors
Xiaoxiao Liu and Mengyuan Chen
Mech. Sci., 14, 293–304, https://doi.org/10.5194/ms-14-293-2023, https://doi.org/10.5194/ms-14-293-2023, 2023
Short summary
Short summary
Considering its practical engineering applications, a proportional-derivative-based robust controller is designed and applied to a nonlinear robotic system with uncertainty. With error- and model-based features, accurate position tracking can be achieved. The control feedback gain is automatically and iteratively adjusted by learning so that the desired performance of the system is optimized.
Xiaoxiao Liu and Mengyuan Chen
Mech. Sci., 14, 293–304, https://doi.org/10.5194/ms-14-293-2023, https://doi.org/10.5194/ms-14-293-2023, 2023
Short summary
Short summary
Considering its practical engineering applications, a proportional-derivative-based robust controller is designed and applied to a nonlinear robotic system with uncertainty. With error- and model-based features, accurate position tracking can be achieved. The control feedback gain is automatically and iteratively adjusted by learning so that the desired performance of the system is optimized.
Related subject area
Subject: Mechanisms and Robotics | Techniques and Approaches: Experiment and Best Practice
Design and performance analysis of the 4UPS-RRR parallel ankle rehabilitation mechanism
Assistance control strategy for upper-limb rehabilitation robot based on motion trend
Experimental study on fingertip friction perception characteristics on ridged surfaces
Wearable ankle assistance robot for a human walking with different loads
A vision-based robotic system following the human upper-limb sewing action
Gait analysis algorithm for lower limb rehabilitation robot applications
Obstacle-avoidance path planning based on the improved artificial potential field for a 5 degrees of freedom bending robot
Development of a force-field-based control strategy for an upper-limb rehabilitation robot
Assessment of force control for surface finishing – an experimental comparison between Universal Robots UR10e and FerRobotics active contact flange
Characteristics of the Received Signal of an Ultrasonic Sensor Installed in a Chamber with Micro-Leakage
Crushing mechanism of a mobile pellet harvester
Design and analysis of an innovative flapping wing micro aerial vehicle with a figure eight wingtip trajectory
Development of a novel flapping wing micro aerial vehicle with elliptical wingtip trajectory
Design of a robot-assisted exoskeleton for passive wrist and forearm rehabilitation
Design and hardware selection for a bicycle simulator
Design and evaluation of a continuum robot with extendable balloons
A novel 5-DOF high-precision compliant parallel mechanism for large-aperture grating tiling
Development of a lower extremity wearable exoskeleton with double compact elastic module: preliminary experiments
Influence of gear loss factor on the power loss prediction
Experimental tests on operation performance of a LARM leg mechanism with 3-DOF parallel architecture
Fatigue testing of flexure hinges for the purpose of the development of a high-precision micro manipulator
An articulated handle to improve the ergonomic performance of robotic dextrous instruments for laparoscopic surgery
Devices for accurate placement of epidural Tuohy needle for Anaesthesia administration
Power-free bistable threshold accelerometer made from a carbon nanotube framework
Origami-like creases in sheet materials for compliant mechanism design
Review article: locomotion systems for ground mobile robots in unstructured environments
Kan Shi, Zongjia Wang, Changtao Yan, and Zhiwei Wang
Mech. Sci., 15, 417–430, https://doi.org/10.5194/ms-15-417-2024, https://doi.org/10.5194/ms-15-417-2024, 2024
Short summary
Short summary
In order to lessen the impact on the team of rehabilitation practitioners and provide patients with a higher-quality rehabilitation process, an ankle rehabilitation robot based on a parallel mechanism is proposed. The feasibility of the ankle rehabilitation robot proposed in this paper is proven by analysis, which lays a foundation for future human–machine experiments. It can act as a reference for future research of the ankle rehabilitation mechanism.
Haojun Zhang, Tao Song, and Leigang Zhang
Mech. Sci., 14, 503–518, https://doi.org/10.5194/ms-14-503-2023, https://doi.org/10.5194/ms-14-503-2023, 2023
Short summary
Short summary
To enable the control system to provide minimal assistance and apply different rehabilitation stages according to the subject's performance, this paper proposes a motion-trend-based assistance control strategy. The preliminary experimental results demonstrate that the proposed control strategy works well to quickly adjust the assistance to the subject's motor performance and quickly reduce the assistance when the subject tends to actively participate in the exercise.
Liyong Wang, Li Yang, Le Li, Jianpeng Wu, and Qian Zou
Mech. Sci., 14, 463–477, https://doi.org/10.5194/ms-14-463-2023, https://doi.org/10.5194/ms-14-463-2023, 2023
Short summary
Short summary
The development of bionic skin has always been a challenging scientific research problem. A novel experimental method is proposed for investigating fingertip friction perception characteristics. The results show that the tactile perception accuracy can be improved by changing the surface texture and lubrication conditions. The method can provide peer experience for revealing tactile perception mechanisms and can also provide theoretical guidance for the research of bionic skin.
Junqiang Li, Kuan Yang, and Dong Yang
Mech. Sci., 14, 429–438, https://doi.org/10.5194/ms-14-429-2023, https://doi.org/10.5194/ms-14-429-2023, 2023
Short summary
Short summary
A wearable ankle assisted robot is developed to meet the demand for ankle assistance during human walking. The active energy storage mechanism in the robot can realize the storage and release of energy and realize the application of a low-power motor to provide large assisted force. The experimental results show that the net metabolic cost of the participants is reduced by averages of 5.30 %, 5.67 %, and 4.84 % with 0, 4, and 8 kg loads respectively.
Liming Zhang, Xiaohua Wang, Haoyi Wang, and Pengfei Li
Mech. Sci., 14, 347–359, https://doi.org/10.5194/ms-14-347-2023, https://doi.org/10.5194/ms-14-347-2023, 2023
Short summary
Short summary
This paper proposes a robot following method to follow the sewing action of human upper limbs. The purpose of this paper is to demonstrate that robots can assist workers in the future. Compared with the visual servo method, the accuracy is much improved.
Li Zheng and Tao Song
Mech. Sci., 14, 315–331, https://doi.org/10.5194/ms-14-315-2023, https://doi.org/10.5194/ms-14-315-2023, 2023
Short summary
Short summary
In this paper, a gait analysis algorithm for a lower limb rehabilitation robot is proposed. The algorithm realizes the division of gait temporal information and the design of a gait spatiotemporal parameter algorithm based on lidar. A spatial parameter-splicing algorithm based on a time series is proposed, which effectively reduces the influence of errors on gait parameters. Based on the gait algorithm, aiming at real-time algorithm performance, a dynamic window method is proposed.
Quansheng Jiang, Kai Cai, and Fengyu Xu
Mech. Sci., 14, 87–97, https://doi.org/10.5194/ms-14-87-2023, https://doi.org/10.5194/ms-14-87-2023, 2023
Short summary
Short summary
Traditional path-planning algorithms for bending robots are often a sequence of multiple line segments and an unsmoothed curve, which causes a discontinuous robot motion. A smooth path can ensure continuous motion. We propose an improved artificial potential field-based path-planning method based on the idea of a rapidly exploring random tree (RRT) algorithm, reducing the length of and smoothing the path, thus solving the path-planning problem of a multi-degrees-of-freedom (DOF) bending robot.
Jiasheng Pan, Leigang Zhang, and Qing Sun
Mech. Sci., 13, 949–959, https://doi.org/10.5194/ms-13-949-2022, https://doi.org/10.5194/ms-13-949-2022, 2022
Short summary
Short summary
Robot-assisted rehabilitation has proven to be effective for improving the motor performance of patients with neuromuscular injuries. Therefore, the main purpose of this paper is to present a new patient-cooperative control framework for an end-effector upper-limb rehabilitation robot that provides robot-assisted training for individuals with neuromuscular disorders. The feasibility of the proposed control scheme is validated via training experiments using five healthy subjects.
Stefan Gadringer, Hubert Gattringer, and Andreas Mueller
Mech. Sci., 13, 361–370, https://doi.org/10.5194/ms-13-361-2022, https://doi.org/10.5194/ms-13-361-2022, 2022
Short summary
Short summary
We introduce different test scenarios and set-ups for a force control assessment. The force control of Universal Robots (UR10e) and FerRobotics (ACF-K 109/04) are evaluated. A force/torque sensor, mounted below a work piece, measures the applied force by the UR or by the ACF. Results for both UR and ACF force control are presented for varying desired contact velocities and forces. These results show the advantage of the ACF-K 109/04 over the UR10e force control for highly dynamic scenarios.
Wonjun Seo, Seokyeon Im, and Geesoo Lee
Mech. Sci., 12, 1051–1060, https://doi.org/10.5194/ms-12-1051-2021, https://doi.org/10.5194/ms-12-1051-2021, 2021
Short summary
Short summary
This study is focused on analyzing signal changes according to internal flow and temperature change by generating micro-leakage, unlike previous studies that experimented with no flow inside a chamber. In addition, a new application method utilizing a phase shift by post-processing the ultrasonic reception signal was proposed. By using the ultrasonic sensor, a different method was proposed for the measurement of micro-leakage of gaseous fuel that relied on the existing pressure sensor.
Yanyan Ge, Zichao Su, Maohua Xiao, Min Kang, Ruyi Wang, and Qin Zeng
Mech. Sci., 12, 725–733, https://doi.org/10.5194/ms-12-725-2021, https://doi.org/10.5194/ms-12-725-2021, 2021
Short summary
Short summary
This study proposed and verified a new idea for the crushing system of a mobile pellet harvester. As the key components of the crushing mechanism, the theoretical design calculation was given, and Ansys software was used to verify the feasibility of the design, specifically by computing the strength and stiffness. Field tests were conducted, and the test results met the requirements. This design can promote the further development of a mobile pellet harvester.
Shan Jiang, Yong Hu, Qiang Li, Long Ma, Yang Wang, Xiaoqin Zhou, and Qiang Liu
Mech. Sci., 12, 603–613, https://doi.org/10.5194/ms-12-603-2021, https://doi.org/10.5194/ms-12-603-2021, 2021
Short summary
Short summary
An innovative flapping wing micro aerial vehicle (FWMAV), forming a figure eight wingtip trajectory, which can achieve complex composite motions of flapping, twisting, and swinging is presented in this paper. Along with the design concept of reducing any possible weight and size, the aircraft was designed with classical and reliable mechanical components. Then, experiments were conducted to test the FWMAV aerodynamic efficiency with a complex figure eight wingtip trajectory.
Qiang Liu, Qiang Li, Xiaoqin Zhou, Pengzi Xu, Luquan Ren, and Shengli Pan
Mech. Sci., 10, 355–362, https://doi.org/10.5194/ms-10-355-2019, https://doi.org/10.5194/ms-10-355-2019, 2019
Mehmet Erkan Kütük, Lale Canan Dülger, and Memik Taylan Daş
Mech. Sci., 10, 107–118, https://doi.org/10.5194/ms-10-107-2019, https://doi.org/10.5194/ms-10-107-2019, 2019
Short summary
Short summary
This paper presents a new exoskeleton design for wrist and forearm rehabilitation. The contribution of this study is to offer a methodology which shows how to adapt a serial manipulator that reduces the number of actuators used in exoskeleton design for the rehabilitation. The system offered is a combination of end-effector- and exoskeleton-based devices. The passive exoskeleton is attached to the end effector of the manipulator, which provides motion for the purpose of rehabilitation process.
Georgios Dialynas, Riender Happee, and Arend L. Schwab
Mech. Sci., 10, 1–10, https://doi.org/10.5194/ms-10-1-2019, https://doi.org/10.5194/ms-10-1-2019, 2019
Short summary
Short summary
With the resurgence in bicycle ridership in the last decade, a better understanding of bicycle rider
behavior is important in accommodating increasing bicycle traffic as well as reducing accidents that come along with it. At TU Delft we have designed a fixed-base bicycle simulator to understand how bicycle rider's interact with other road users in a safe virtual environment.
Efe Yamac Yarbasi and Evren Samur
Mech. Sci., 9, 51–60, https://doi.org/10.5194/ms-9-51-2018, https://doi.org/10.5194/ms-9-51-2018, 2018
Short summary
Short summary
This article presents a novel continuum robot actuated by two extendable balloons. As inflated, the balloons apply a force on the wall of the tip, pushing the robot forward. The contribution of this study is the introduction of a novel actuation mechanism for soft robots to have extreme elongation (2000 %) in order to be navigated in substantially long and narrow environments.
Zhongxi Shao, Shilei Wu, Jinguo Wu, and Hongya Fu
Mech. Sci., 8, 349–358, https://doi.org/10.5194/ms-8-349-2017, https://doi.org/10.5194/ms-8-349-2017, 2017
Yi Long, Zhi-jiang Du, Chao-feng Chen, Wei-dong Wang, and Wei Dong
Mech. Sci., 8, 249–258, https://doi.org/10.5194/ms-8-249-2017, https://doi.org/10.5194/ms-8-249-2017, 2017
Short summary
Short summary
Compared to the traditional pHRI measurement approaches, the proposed method arranged the sensors in the mechanical joint instead of the connection cuff. This kind of architecture has compact architecture and improves the wearing comfort, which can adapt to various operators and is convenient to be applied in the wearable exoskeleton. The performance of the DEM will be studied in the following work and human-exoskeleton coordination control strategies will be investigated in future work.
C. M. C. G. Fernandes, P. M. T. Marques, R. C. Martins, and J. H. O. Seabra
Mech. Sci., 6, 81–88, https://doi.org/10.5194/ms-6-81-2015, https://doi.org/10.5194/ms-6-81-2015, 2015
Short summary
Short summary
The importance of the gear loss factor in the power loss predictions was put in evidence comparing the predictions with experimental results.
It was concluded that the gear loss factor is a decisive factor to accurately predict the power loss.
Different formulations proposed in the literature were compared and it was shown that only few were able to yield satisfactory correlations with experimental results.
M. F. Wang, M. Ceccarelli, and G. Carbone
Mech. Sci., 6, 1–8, https://doi.org/10.5194/ms-6-1-2015, https://doi.org/10.5194/ms-6-1-2015, 2015
Short summary
Short summary
In this paper, 1) We design and built a prototype of a LARM tripod leg mechanism with low-cost easy-operation features that can be useful in a biped locomotor; 2) We present an experimental layout for investigating operational performance of the built prototype; 3) We characterize and analyze the system behaviour by follow a prescribed step movement of the foot platform under consideration with human walking.
I. Ivanov and B. Corves
Mech. Sci., 5, 59–66, https://doi.org/10.5194/ms-5-59-2014, https://doi.org/10.5194/ms-5-59-2014, 2014
B. Herman, A. Devreker, F. Richer, A. Hassan Zahraee, and J. Szewczyk
Mech. Sci., 5, 21–28, https://doi.org/10.5194/ms-5-21-2014, https://doi.org/10.5194/ms-5-21-2014, 2014
N. Vaughan, V. N. Dubey, M. Y. K. Wee, and R. Isaacs
Mech. Sci., 5, 1–6, https://doi.org/10.5194/ms-5-1-2014, https://doi.org/10.5194/ms-5-1-2014, 2014
J. D. Tanner and B. D. Jensen
Mech. Sci., 4, 397–405, https://doi.org/10.5194/ms-4-397-2013, https://doi.org/10.5194/ms-4-397-2013, 2013
K. C. Francis, J. E. Blanch, S. P. Magleby, and L. L. Howell
Mech. Sci., 4, 371–380, https://doi.org/10.5194/ms-4-371-2013, https://doi.org/10.5194/ms-4-371-2013, 2013
L. Bruzzone and G. Quaglia
Mech. Sci., 3, 49–62, https://doi.org/10.5194/ms-3-49-2012, https://doi.org/10.5194/ms-3-49-2012, 2012
Cited articles
An, L., Pan, X., Li, T., and Wang, M.: A visual dynamic-SLAM method based semantic segmentation and multi-view geometry, in: Proceedings of the International Conference on High Performance Computing and Communication, Xiamen, China, 3–5 December 2021, 255–263, https://doi.org/10.1117/12.2628175, 2022.
Barnes, C., Shechtman, E., Finkelstein, A., and Goldman, D. B.: PatchMatch: A randomized correspondence algorithm for structural image editing, ACM T. Graphic., 28, 10 pp., 2009.
Bescos, B., Fácil, J. M., Civera, J., and Neira, J.: DynaSLAM: Tracking, mapping, and inpainting in dynamic scenes, IEEE Robotics and Automation Letters, 3, 4076–4083, https://doi.org/10.1109/LRA.2018.2860039, 2018.
Campos, C., Elvira, R., Rodríguez, J. J. G., Montiel, J. M. M., and Tardós J. D.: Orb-slam3: An accurate open-source library for visual, visual–inertial, and multimap slam, IEEE T. Robot., 37, 1874–1890, https://doi.org/10.1109/TRO.2021.3075644, 2021.
Cao, J., Yu, J., Pan, S., Gao, F., Yu, C., Xu, Z., Huang, Z., and Wang, Y.: SLAM pose graph optimization method using dual visual odometry, Journal of Computer Aided Design and Graphics, 33, 1264–1272, 2021.
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., and Zagoruyko, S.: End-to-end object detection with transformers, in: Proceedings of the 2020 European conference on computer vision, 2020 European conference on computer vision workshops, Glasgow, 23–28 August 2020, Springer, 213–229, 2020.
Chen, L. C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H.: Encoder-decoder with atrous separable convolution for semantic image segmentation, in: Proceedings of the European conference on computer vision, European conference on computer vision workshops, Munich, Germany, 8–14 September 2018, Springer, 801–818, 2018.
Chen, M., Ding, L., and Zhang, Y.: Fast PL-SLAM Algorithm Based on Improved Keyframe Selection Strategy, Acta Electronica Sinica, 50, 608–618, https://doi.org/10.12263/DZXB.20210567, 2022.
Chen, M., Guo, H., Qian, R., Gong, G., and Cheng, H.: VSLAM Algorithm Based on Improved Vision Transformer Semantic Segmentation in Dynamic Scenes, GitHub [code/video], https://github.com/Oneghr/VTD-SLAM (last access: 19 December 2023), 2023.
Dai, J., Li, Y., He, K., and Sun, J.: R-fcn: Object detection via region-based fully convolutional networks, GitHub [code], https://github.com/daijifeng001/r-fcn (last access: 19 December 2023), 2016.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N.: An image is worth 16x16 words: Transformers for image recognition at scale, arXiv[preprint], https://doi.org/10.48550/arXiv.2010.11929, 3 June 2021.
Feng, X., Pei, W., Jia, Z., Chen, F., Zhang, D., and Lu, G.: Deep-masking generative network: A unified framework for background restoration from superimposed images, IEEE T. Image Process., 30, 4867–4882, https://doi.org/10.1109/TIP.2021.3076589, 2021.
Gao, X., Shi, X., Ge, Q., and Chen, K.: A review of visual SLAM for dynamic object scenes, Robotics, 43, 733–750, https://doi.org/10.13973/j.cnki.robot.200323, 2021.
Guan, L., Franco, J. S., and Pollefeys, M.: Multi-view occlusion reasoning for probabilistic silhouette-based dynamic scene reconstruction, Int. J. Comput. Vision, 90, 283–303, https://doi.org/10.1007/s11263-010-0341-y, 2010.
Han, K., Wang, Y., Chen, H., Chen, X., Guo, J., Liu, Z., Tang, Y., Xiao, A., Xu, C., Xu, Y., Yang, Z., Zhang, Y., and Tao, D.: A survey on vision transformer, IEEE T. Pattern Anal., 45, 87–110, https://doi.org/10.1109/TPAMI.2022.3152247, 2022.
Hu, J., Shen, L., and Sun, G.: Squeeze-and-excitation networks, in: Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, IEEE conference on computer vision and pattern recognition workshops, Salt Lake City, UT, USA, 18–22 June 2018, IEEE, 7132–7141, 2018.
Huang, B., Zhao, J., and Liu, J.: A survey of simultaneous localization and mapping with an Envision in 6G Wireless Networks, arXiv [preprint], https://doi.org/10.48550/arXiv.1909.05214, 14 February, 2020.
Kaneko, M., Iwami, K., Ogawa, T., Yamasaki, T., and Aizawa, K.: Mask-slam: Robust feature-based monocular slam by masking using semantic segmentation, in: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Salt Lake City, USA, 18–22 June 2018, IEEE, 258–266, 2018.
Lin, T. Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C. L.: Microsoft coco: Common objects in context, in: 13th European Conference on Computer Vision, European Conference on Computer Vision workshops, Zurich, Switzerland, 6–12 September 2014, Springer, 8693, 740–755, 2014.
Liu, Y. and Miura, J.: RDS-SLAM: real-time dynamic SLAM using semantic segmentation methods, IEEE Access, 9, 23772–23785, https://doi.org/10.1109/ACCESS.2021.3050617, 2021.
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows, in: Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Virtual, 11–17 October 2021, IEEE, 10012—10022, 2021.
Liu, Z., Yang, D., Wang, Y., Liu, M., and Li, R.: EGNN: Graph structure learning based on evolutionary computation helps more in graph neural networks, Appl. Soft Comput., 135, 110040, https://doi.org/10.1016/j.asoc.2023.110040, 2023.
Mur, A. R. and Tardós, J. D.: Orb-slam2: An open-source slam system for monocular, stereo, and rgb-d cameras, IEEE T. Robot., 33, 1255–1262, https://doi.org/10.1109/TRO.2017.2705103, 2017.
Pumarola, A., Vakhitov, A., Agudo, A., Sanfeliu, A., and Moreno, Nr. F.: PL-SLAM: Real-time monocular visual SLAM with points and lines, in: Proceedings of the 2017 IEEE International Conference on Robotics and Automation (ICRA), Singapore, 29 May–3 June 2017, IEEE, 4503–4508, https://doi.org/10.1109/ICRA.2017.7989522, 2017.
Schubert, D., Goll, T., Demmel, N., Usenko, V., Stückler, J., and Cremers, D.: The TUM VI benchmark for evaluating visual-inertial odometry, in: Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 1–5 October 2018, IEEE, 1680–1687, https://doi.org/10.1109/IROS.2018.8593419, 2018.
Shi, Y., Li, H., Fu, X., Luan, R., Wang, Y., Sun, Z., Niu, Y., Wang, C., Zhang, C., and Wang, Z. L.: Self-powered difunctional sensors based on sliding contact-electrification and tribovoltaic effects for pneumatic monitoring and controlling, Nano Energy, 110, 108339, https://doi.org/10.1016/j.nanoen.2023.108339, 2023a.
Shi, Y., Li, L., Yang, J., Wang, Y., and Hao, S.: Center-based transfer feature learning with classifier adaptation for surface defect recognition, Mech. Syst. Signal Pr., 188, 110001, https://doi.org/10.1016/j.ymssp.2022.110001, 2023b.
Strudel, R., Garcia, R., Laptev, I., and Schmid, C.: Segmenter: Transformer for semantic segmentation, in: Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Virtual, 11–17 October 2021, IEEE, 597–606, 2021
Sun, Y., Liu, M., and Meng, M.: Motion removal for reliable RGB-D SLAM in dynamic environments, Robot. Auton. Syst., 108, 115–128, https://doi.org/10.1016/j.robot.2018.07.002, 2018.
Targ, S., Almeida, D., and Lyman, K.: Resnet in resnet: Generalizing residual architectures, arXiv [preprint], https://doi.org/10.48550/arXiv.1603.08029, 25 March 2016.
Tian, C., Xu, Z., Wang, L., and Liu, Y.: Arc fault detection using artificial intelligence: Challenges and benefits, Math. Biosci. Eng., 20, 12404–12432, https://doi.org/10.3934/mbe.2023552, 2023.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I.: Attention is all you need, Adv. Neur. In., 30, 11 pp., 2017.
Wang, Y., Liu, Z., Xu, J., and Yan, W.: Heterogeneous network representation learning approach for ethereum identity identification, IEEE Transactions on Computational Social Systems, 10, 890 –899, https://doi.org/10.1109/TCSS.2022.3164719, 2023.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I.: Attention is all you need, in: Proceedings of the Annual Conference on Neural Information Processing Systems, Conference and Workshop on Neural Information Processing Systems workshops, Long Beach, Los Angeles area, USA, 4–9 December 2017, MIT Press, Los Angeles, ISBN 9781510860964, 2017.
Woo, S., Park, J., Lee, J. Y., and Kweon, I. S.: Cbam: Convolutional block attention module, in: Proceedings of the European conference on computer vision, European conference on computer vision workshops, Munich, Germany, 8–14 September 2018, 3–19, 2018.
Xu, C., Zhou, Y., and Luo, C.: Visual SLAM Method Based on Optical Flow and Instance Segmentation in Dynamic Scenes, Acta Optica Sinica, 42, 147–159, https://doi.org/10.3788/AOS202242.1415002, 2022.
Yu, C., Liu, Z., Liu, X. J., Xie, F., Yang, Y., Wei, Q., and Fei, Q.: DS-SLAM: A semantic visual SLAM towards dynamic environments, in: Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IEEE/RSJ International Conference on Intelligent Robots and Systems workshops, Madrid, Spain, 1–5 October 2018, IEEE, 1168–1174, https://doi.org/10.1109/IROS.2018.8593691, 2018.
Zhao, J. and Lv, Y.: Output-feedback Robust Tracking Control of Uncertain Systems via Adaptive Learning, Int. J. Control. Autom., 21, 1108–1118, https://doi.org/10.1007/s12555-021-0882-6, 2023.
Zhong, F., Wang, S., Zhang, Z., Chen, C., and Wang, Y.: Detect-SLAM: Making object detection and SLAM mutually beneficial, in: Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision, Harrah's and Harvey's Lake Tahoe, NV, USA, 11–15 March 2018, IEEE, 1001–1010, https://doi.org/10.1109/WACV.2018.00115, 2018.
Zhou, Z., Cao, J., and Di, S.: A review of SLAM algorithm for 3D lidar, Journal of Instrument and Instrument, 42, 13–27, https://doi.org/10.19650/j.cnki.cjsi.J2107897, 2021.
Short summary
The proposed VTD-SLAM algorithm has the following advantages. (1) A multiclass feature enhancement and multiclass feature guidance semantic segmentation network (MSNET) is proposed to improve the semantic segmentation ability of dynamic objects. (2) Optimal neighbor pixel matching is used to complete the image of the deleted region. Our method can effectively solve the influence of dynamic objects, so that the SLAM system can operate effectively.
The proposed VTD-SLAM algorithm has the following advantages. (1) A multiclass feature...
