Zheng Yao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zheng Yao has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Hanxing Xie
Beihang University
1 shared publication - Huixia Wang
Beijing Aerospace Automatic Control Institute
1 shared publication - Jiayu She
Department of Aerospace Engineering, Harbin Institute of Technology, Harbin, China
1 shared publication - Kunfeng Lu
Beijing Aerospace Automatic Control Institute
1 shared publication - Naiming Qi
Harbin Institute of Technology
1 shared publication - Rui Zhou
School of Astronautics, Harbin Institute of Technology, Harbin 150001, China
1 shared publication
Latest Publications
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 11038-11049
STDSD: A Spacecraft Target Detection Framework Considering Similarity and Diversity
Benefiting from the development of deep learning and computer vision, object detection performance has witnessed remarkable improvements in various domains recently. Nonetheless, several challenges persist for spacecraft detection, including challenging space environment, domain gaps, and real-time application. This paper proposes a Spacecraft Target Detection framework considering Similarity and Diversity strategy (STDSD), specifically designed to adapt and excel in the special requirements of space object detection. The STDSD framework integrates multiple feature extraction networks (FNets) and overlapping vectors prediction networks (ONets). Initially, the original images are cropped into several segments, which are processed through the FNets to extract critical features. These features are subsequently input to the ONets, where adversarial learning and random dropout techniques are implemented to enhance network diversity. This strategy of incorporating both similarity and diversity significantly enhances the precision and robustness of target detection, especially in bridging domain gaps. STDSD was evaluated against the existing detectors using the SPEED, SPEED+, SwissCube, Web-scraped, and Hardware-in-loop datasets. Specifically, STDSD achieved an F-score of 0.9989 on the SPEED dataset, 0.9876 on the “Lightbox” subset, and 0.9523 on the “Sunlamp” subset of the SPEED+ dataset. Additionally, on the SwissCube, Web-scraped, and Hardware-in-loop datasets, the F-scores were 0.9805, 0.8848, and 0.9879, respectively. Furthermore, STDSD exhibited a remarkable capability of over 10 frames per second (FPS) @GPU on a laptop. These results demonstrate that STDSD achieves high prediction accuracy and processing speed, underscoring the robustness capabilities across different datasets and the practicality for real-time applications in space environment.
- 23 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 22 (15455955) , 11038-11049
STDSD: A Spacecraft Target Detection Framework Considering Similarity and Diversity
Benefiting from the development of deep learning and computer vision, object detection performance has witnessed remarkable improvements in various domains recently. Nonetheless, several challenges persist for spacecraft detection, including challenging space environment, domain gaps, and real-time application. This paper proposes a Spacecraft Target Detection framework considering Similarity and Diversity strategy (STDSD), specifically designed to adapt and excel in the special requirements of space object detection. The STDSD framework integrates multiple feature extraction networks (FNets) and overlapping vectors prediction networks (ONets). Initially, the original images are cropped into several segments, which are processed through the FNets to extract critical features. These features are subsequently input to the ONets, where adversarial learning and random dropout techniques are implemented to enhance network diversity. This strategy of incorporating both similarity and diversity significantly enhances the precision and robustness of target detection, especially in bridging domain gaps. STDSD was evaluated against the existing detectors using the SPEED, SPEED+, SwissCube, Web-scraped, and Hardware-in-loop datasets. Specifically, STDSD achieved an F-score of 0.9989 on the SPEED dataset, 0.9876 on the “Lightbox” subset, and 0.9523 on the “Sunlamp” subset of the SPEED+ dataset. Additionally, on the SwissCube, Web-scraped, and Hardware-in-loop datasets, the F-scores were 0.9805, 0.8848, and 0.9879, respectively. Furthermore, STDSD exhibited a remarkable capability of over 10 frames per second (FPS) @GPU on a laptop. These results demonstrate that STDSD achieves high prediction accuracy and processing speed, underscoring the robustness capabilities across different datasets and the practicality for real-time applications in space environment.
- Open Access
Onboard Distributed Trajectory Planning through Intelligent Search for Multi-UAV Cooperative Flight
Trajectory planning and obstacle avoidance play essential roles in the cooperative flight of multiple unmanned aerial vehicles (UAVs). In this paper, a unified framework for onboard distributed trajectory planning is proposed, which takes full advantage of intelligent discrete and continuous search algorithms. Firstly, the Monte Carlo tree search (MCTS) is used as the task allocation algorithm to solve the cooperative obstacle avoidance problem. Taking the task allocation decisions as the constraint, knowledge-based particle swarm optimization (Know-PSO) is used as the optimization algorithm to solve the onboard distributed cooperative trajectory planning problem. Simulation results demonstrate that the proposed intelligent MCTS-PSO search framework is effective and flexible for multiple UAVs to conduct the cooperative trajectory planning and obstacle avoidance. Further, it has been applied in practical experiments and achieved promising results.
Open AccessOnboard Distributed Trajectory Planning through Intelligent Search for Multi-UAV Cooperative Flight
Trajectory planning and obstacle avoidance play essential roles in the cooperative flight of multiple unmanned aerial vehicles (UAVs). In this paper, a unified framework for onboard distributed trajectory planning is proposed, which takes full advantage of intelligent discrete and continuous search algorithms. Firstly, the Monte Carlo tree search (MCTS) is used as the task allocation algorithm to solve the cooperative obstacle avoidance problem. Taking the task allocation decisions as the constraint, knowledge-based particle swarm optimization (Know-PSO) is used as the optimization algorithm to solve the onboard distributed cooperative trajectory planning problem. Simulation results demonstrate that the proposed intelligent MCTS-PSO search framework is effective and flexible for multiple UAVs to conduct the cooperative trajectory planning and obstacle avoidance. Further, it has been applied in practical experiments and achieved promising results.
- Open Access
- 18 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 40991-41005
Intermittent Gliding Flight Control Design and Verification of a Morphing Unmanned Aerial Vehicle
The intermittent gliding flight method taking advantage of the updraft is adopted to increase the flight endurance time of a type of multifunctional morphing unmanned aerial vehicle (UAV). This type of UAV is able to be folded before launch, launch from a cylindrical launch tube and adjust the wing span and sweep angle during flight. It has the advantages of convenient storage and transportation and can use the energy of the updraft. In order to effectively utilize the updraft, the mathematical model of updraft is established, and an updraft judging method is proposed. In order to complete the intermittent gliding flight, the flight progress is divided into two stages and a segmentation control law is proposed. During the intermittent gliding stage switching, the wing sweep angle and wing span length are varied to accommodate requirements of different stages. H robust control design method for gain scheduling based on linear parameter varying (LPV) model is used to ensure the stability of the morphing UAV during the wings change. Mathematical simulation verifies the effectiveness of the morphing control algorithm. After that, a semi-physical simulation verification involving a mathematical UAV model with a physical wing changing mechanism, a physical embedded flight control module, the real flight attitude and the virtual route are implemented. It has been verified that the morphing UAV can remain stable throughout the entire cycle of the intermittent gliding. Moreover, the intermittent gliding flight with updraft can reduce the fuel consumption of the morphing UAV and increase its flight endurance time.
- 18 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 40991-41005
Open AccessIntermittent Gliding Flight Control Design and Verification of a Morphing Unmanned Aerial Vehicle
The intermittent gliding flight method taking advantage of the updraft is adopted to increase the flight endurance time of a type of multifunctional morphing unmanned aerial vehicle (UAV). This type of UAV is able to be folded before launch, launch from a cylindrical launch tube and adjust the wing span and sweep angle during flight. It has the advantages of convenient storage and transportation and can use the energy of the updraft. In order to effectively utilize the updraft, the mathematical model of updraft is established, and an updraft judging method is proposed. In order to complete the intermittent gliding flight, the flight progress is divided into two stages and a segmentation control law is proposed. During the intermittent gliding stage switching, the wing sweep angle and wing span length are varied to accommodate requirements of different stages. H robust control design method for gain scheduling based on linear parameter varying (LPV) model is used to ensure the stability of the morphing UAV during the wings change. Mathematical simulation verifies the effectiveness of the morphing control algorithm. After that, a semi-physical simulation verification involving a mathematical UAV model with a physical wing changing mechanism, a physical embedded flight control module, the real flight attitude and the virtual route are implemented. It has been verified that the morphing UAV can remain stable throughout the entire cycle of the intermittent gliding. Moreover, the intermittent gliding flight with updraft can reduce the fuel consumption of the morphing UAV and increase its flight endurance time.
- 17 May 2014
- Published by Elsevier in Chinese Journal of Mechanical Engineering
- Vol. 27 (3) , 437-447
Proceeding of human exoskeleton technology and discussions on future research
After more than half a century of intense efforts, the development of exoskeleton has seen major advances, and several remarkable achievements have been made. Reviews of developing history of exoskeleton are presented, both in active and passive categories. Major models are introduced, and typical technologies are commented on. Difficulties in control algorithm, driver system, power source, and man-machine interface are discussed. Current researching routes and major developing methods are mapped and critically analyzed, and in the process, some key problems are revealed. First, the exoskeleton is totally different from biped robot, and relative studies based on the robot technologies are considerably incorrect. Second, biomechanical studies are only used to track the motion of the human body, the interaction between human and machines are seldom studied. Third, the traditional developing ways which focused on servo-controlling have inborn deficiency from making portable systems. Research attention should be shifted to the human side of the coupling system, and the human ability to learn and adapt should play a more significant role in the control algorithms. Having summarized the major difficulties, possible future works are discussed. It is argued that, since a distinct boundary cannot be drawn in such strong-coupling human-exoskeleton system, the more complex the control system gets, the more difficult it is for the user to learn to use. It is suggested that the exoskeleton should be treated as a simple wearable tool, and downgrading its automatic level may be a change toward a brighter research outlook. This effort at simplification is definitely not easy, as it necessitates theoretical supports from fields such as biomechanics, ergonomics, and bionics.
- 17 May 2014
- Published by Elsevier in Chinese Journal of Mechanical Engineering
- Vol. 27 (3) , 437-447
Proceeding of human exoskeleton technology and discussions on future research
After more than half a century of intense efforts, the development of exoskeleton has seen major advances, and several remarkable achievements have been made. Reviews of developing history of exoskeleton are presented, both in active and passive categories. Major models are introduced, and typical technologies are commented on. Difficulties in control algorithm, driver system, power source, and man-machine interface are discussed. Current researching routes and major developing methods are mapped and critically analyzed, and in the process, some key problems are revealed. First, the exoskeleton is totally different from biped robot, and relative studies based on the robot technologies are considerably incorrect. Second, biomechanical studies are only used to track the motion of the human body, the interaction between human and machines are seldom studied. Third, the traditional developing ways which focused on servo-controlling have inborn deficiency from making portable systems. Research attention should be shifted to the human side of the coupling system, and the human ability to learn and adapt should play a more significant role in the control algorithms. Having summarized the major difficulties, possible future works are discussed. It is argued that, since a distinct boundary cannot be drawn in such strong-coupling human-exoskeleton system, the more complex the control system gets, the more difficult it is for the user to learn to use. It is suggested that the exoskeleton should be treated as a simple wearable tool, and downgrading its automatic level may be a change toward a brighter research outlook. This effort at simplification is definitely not easy, as it necessitates theoretical supports from fields such as biomechanics, ergonomics, and bionics.