Yinlong Yuan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yinlong Yuan has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Yun Cheng
School of Electrical Engineering and Automation, Nantong University, Nantong 226019, China
8 shared publications - Liang Hua
School of Electrical Engineering and Automation, Nantong University, Nantong 226019, China
7 shared publications Yun Cheng
School of Electrical Engineering and Automation, Nantong University, Nantong, China
4 shared publications- Feng Chen
School of Electrical Engineering and Automation, Nantong University, Nantong, China
2 shared publications - Jin Hu
School of Electrical Engineering and Automation, Nantong University, Nantong, China
2 shared publications - Junhong Li
School of Electrical Engineering and Automation Nantong University Nantong China
2 shared publications
Latest Publications
- 22 October 2026
- Published by Springer Nature in Arabian Journal for Science and Engineering
- p. 1-15
Multi-UAV Autonomous Cooperative Control Based on MASAC and MPC
- 22 October 2026
- Published by Springer Nature in Arabian Journal for Science and Engineering
- p. 1-15
Multi-UAV Autonomous Cooperative Control Based on MASAC and MPC
- Open Access
Multi-History-Weighted Projection-Adaptive Jacobian Control for Close-Range UAV Visual Servoing near Overhead Ground Wires
Close-range UAV visual servoing near overhead ground wires requires simultaneous regulation of the target’s lateral position, apparent width, and orientation angle in the image. Owing to flight-control response lag and visual-processing delay, current image changes may reflect the combined effects of multiple historical control inputs. To address this issue, this paper proposes a multi-history-weighted projection-adaptive Jacobian control method (MHW-PAJ-CLF). For each visual channel, the method takes a weighted sum of the parameter corrections associated with different historical inputs based on response prediction errors and input magnitudes and updates the local Jacobian matrix online under projection bounds. The online estimate is then blended with the nominal model. Control Lyapunov function-based quadratic programming (CLF-QP) uses the blended model to generate velocity and yaw-rate commands, coordinating the regulation of the three visual errors under input constraints. Comparative experiments were conducted on the RflySim–PX4 hardware-in-the-loop platform. The proposed method achieves a mean success rate of 75.50%, exceeding that of the best baseline by 11.05 percentage points. In tests with additional visual-feedback delay and command–response lag, the proposed method achieves lower overall tracking error than single-history PAJ-CLF, indicating that multi-history weighting helps improve visual tracking under time delays.
Open AccessMulti-History-Weighted Projection-Adaptive Jacobian Control for Close-Range UAV Visual Servoing near Overhead Ground Wires
Close-range UAV visual servoing near overhead ground wires requires simultaneous regulation of the target’s lateral position, apparent width, and orientation angle in the image. Owing to flight-control response lag and visual-processing delay, current image changes may reflect the combined effects of multiple historical control inputs. To address this issue, this paper proposes a multi-history-weighted projection-adaptive Jacobian control method (MHW-PAJ-CLF). For each visual channel, the method takes a weighted sum of the parameter corrections associated with different historical inputs based on response prediction errors and input magnitudes and updates the local Jacobian matrix online under projection bounds. The online estimate is then blended with the nominal model. Control Lyapunov function-based quadratic programming (CLF-QP) uses the blended model to generate velocity and yaw-rate commands, coordinating the regulation of the three visual errors under input constraints. Comparative experiments were conducted on the RflySim–PX4 hardware-in-the-loop platform. The proposed method achieves a mean success rate of 75.50%, exceeding that of the best baseline by 11.05 percentage points. In tests with additional visual-feedback delay and command–response lag, the proposed method achieves lower overall tracking error than single-history PAJ-CLF, indicating that multi-history weighting helps improve visual tracking under time delays.
- Open Access
A Geometry-Conditioned Symmetry-Aware Domain-Robust Observation Correction Front-End for Anti-UAV Visual Perception
Although the bounding boxes produced by an object detector provide real-time target localization cues for anti-UAV visual perception, they remain susceptible to geometric deviations under long-range small-target conditions, complex backgrounds, motion blur, and cross-domain environmental variations. Consequently, these detector outputs cannot always serve directly as stable observations for state estimation, trajectory prediction, and interception control. To address this issue, this paper proposes CDBR-Net, a causally conditioned domain-robust observation correction network for post-detection UAV bounding-box refinement. CDBR-Net employs a shared encoder, disentangled multi-branch representations, and a quality-aware gating mechanism to jointly produce a corrected observation box, a robust representation, and an observation uncertainty estimate. To preserve this conditional environment-transformation symmetry without suppressing geometry-induced symmetry breaking, CDBR-Net constructs geometry-conditioned cross-domain sample pairs and imposes cross-domain consistency and geometry-sensitivity preservation constraints. After training on 8749 post-detection observations, CDBR-Net is evaluated on 997 aligned validation observations from three simulated scene domains. It reduces the YOLO bounding-box mean absolute error (MAE) by 6.8%, from 0.002924 to 0.002725, and increases the intersection over union (IoU) by 0.007979, from 0.852167 to 0.860146. It further reduces MAE by 1.7% and increases IoU by 0.002031 relative to the YOLO + MLP Residual baseline. Ablation studies demonstrate the complementary roles of geometry-conditioned cross-domain consistency and geometry-sensitivity preservation. These results indicate that CDBR-Net provides a more stable and geometrically consistent post-detection observation interface for anti-UAV visual perception.
Open AccessA Geometry-Conditioned Symmetry-Aware Domain-Robust Observation Correction Front-End for Anti-UAV Visual Perception
Although the bounding boxes produced by an object detector provide real-time target localization cues for anti-UAV visual perception, they remain susceptible to geometric deviations under long-range small-target conditions, complex backgrounds, motion blur, and cross-domain environmental variations. Consequently, these detector outputs cannot always serve directly as stable observations for state estimation, trajectory prediction, and interception control. To address this issue, this paper proposes CDBR-Net, a causally conditioned domain-robust observation correction network for post-detection UAV bounding-box refinement. CDBR-Net employs a shared encoder, disentangled multi-branch representations, and a quality-aware gating mechanism to jointly produce a corrected observation box, a robust representation, and an observation uncertainty estimate. To preserve this conditional environment-transformation symmetry without suppressing geometry-induced symmetry breaking, CDBR-Net constructs geometry-conditioned cross-domain sample pairs and imposes cross-domain consistency and geometry-sensitivity preservation constraints. After training on 8749 post-detection observations, CDBR-Net is evaluated on 997 aligned validation observations from three simulated scene domains. It reduces the YOLO bounding-box mean absolute error (MAE) by 6.8%, from 0.002924 to 0.002725, and increases the intersection over union (IoU) by 0.007979, from 0.852167 to 0.860146. It further reduces MAE by 1.7% and increases IoU by 0.002031 relative to the YOLO + MLP Residual baseline. Ablation studies demonstrate the complementary roles of geometry-conditioned cross-domain consistency and geometry-sensitivity preservation. These results indicate that CDBR-Net provides a more stable and geometrically consistent post-detection observation interface for anti-UAV visual perception.
- Open Access
- 25 February 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (5) , 425
Autonomous Navigation of an Unmanned Underwater Vehicle via Safe Reinforcement Learning and Active Disturbance Rejection Control
A two-layer control framework for unmanned underwater vehicle (UUV) navigation is proposed, combining a lower-layer active disturbance rejection controller (ADRC) with an upper-layer safe reinforcement learning (RL) policy for obstacle-avoidance navigation. The lower layer, utilizing ADRC, ensures high tracking accuracy and effective disturbance rejection, while the upper layer integrates the twin delayed deep deterministic policy gradient (TD3) algorithm, combined with a control barrier function (CBF)-based quadratic programming (QP) safety filter and safety-inspired reward shaping (SR). The method is evaluated in two simulation studies: (i) velocity and attitude control to assess tracking and disturbance rejection, and (ii) obstacle-avoidance navigation to assess learning efficiency, trajectory smoothness, and safety-related metrics. Simulation results show that ADRC achieves faster tracking and stronger disturbance rejection than a conventional proportional–integral–derivative (PID) controller. Moreover, the proposed TD3 + QP + SR scheme exhibits faster learning, smoother trajectories, and improved safety performance compared with RL baselines. These results indicate that the proposed framework enables efficient and safe UUV navigation in simulation scenarios with obstacles and disturbances.
- 25 February 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (5) , 425
Open AccessAutonomous Navigation of an Unmanned Underwater Vehicle via Safe Reinforcement Learning and Active Disturbance Rejection Control
A two-layer control framework for unmanned underwater vehicle (UUV) navigation is proposed, combining a lower-layer active disturbance rejection controller (ADRC) with an upper-layer safe reinforcement learning (RL) policy for obstacle-avoidance navigation. The lower layer, utilizing ADRC, ensures high tracking accuracy and effective disturbance rejection, while the upper layer integrates the twin delayed deep deterministic policy gradient (TD3) algorithm, combined with a control barrier function (CBF)-based quadratic programming (QP) safety filter and safety-inspired reward shaping (SR). The method is evaluated in two simulation studies: (i) velocity and attitude control to assess tracking and disturbance rejection, and (ii) obstacle-avoidance navigation to assess learning efficiency, trajectory smoothness, and safety-related metrics. Simulation results show that ADRC achieves faster tracking and stronger disturbance rejection than a conventional proportional–integral–derivative (PID) controller. Moreover, the proposed TD3 + QP + SR scheme exhibits faster learning, smoother trajectories, and improved safety performance compared with RL baselines. These results indicate that the proposed framework enables efficient and safe UUV navigation in simulation scenarios with obstacles and disturbances.
- 3 January 2026
- Published by Springer Nature
- p. 307-316
Autonomous Landing Control Method for UAV Multistate Platform Based on Dynamic Weights PPO
- 3 January 2026
- Published by Springer Nature
- p. 307-316
Autonomous Landing Control Method for UAV Multistate Platform Based on Dynamic Weights PPO
- 12 December 2025
- Published by Springer Nature in Arabian Journal for Science and Engineering
- Vol. 51 (9) , 12127-12145
Human–Machine Hybrid Intelligence for Operational Intent Judgment: A Causal Inference Approach
- 12 December 2025
- Published by Springer Nature in Arabian Journal for Science and Engineering
- Vol. 51 (9) , 12127-12145
Human–Machine Hybrid Intelligence for Operational Intent Judgment: A Causal Inference Approach
- 31 October 2025
- Published by Springer Nature in Multimedia Systems
- Vol. 31 (6) , 1-21
A human–machine hybrid intelligence method based on causal representation for solving non-independent and identically distributed problems
- 31 October 2025
- Published by Springer Nature in Multimedia Systems
- Vol. 31 (6) , 1-21
A human–machine hybrid intelligence method based on causal representation for solving non-independent and identically distributed problems
- Open Access
- 1 July 2025
- Published by China Science Publishing & Media Ltd. in JOURNAL OF SHENZHEN UNIVERSITY SCIENCE AND ENGINEERING
- Vol. 42 (4) , 437-446
Maneuver strategy recognition technology for enemy combat aircraft based on Bayesian deep learning
Enhancing identification of enemy combat aircraft maneuver strategies is a critical factor in improving air combat decision-making capabilities. As traditional deep learning models often show overconfidence in complex and variable combat environment, and it is difficult to evaluate the uncertainty, a combat aircraft maneuver strategy recognition method based on Bayesian deep learning is proposed. The method employs Bayesian variational inference and multivariate Gaussian distribution to construct a multi-layer perceptron-based Bayesian deep learning (BDL-MLP) probabilistic model. A gradient balancing factor is introduced to reduce the imbalance between complexity cost gradient and likelihood cost gradient, and then Bayes backpropagation algorithm is used for model training and parameter optimization. Based on the virtual air combat simulation platform AirFlightSim developed in Unity3D software, BDL-MLP, multilayer perceptron (MLP), AlexNet and LeNet models are used to classify and evaluate the data sets of combat aircraft motion scenes with varying degrees of blur (blur radii of 0, 15, 31, 45, and 61 pixels, respectively). Results show that, on the five datasets constructed with the aforementioned blur radii, the maneuver strategy identification accuracy of the BDL-MLP model showed average improvements of 0.43%, 0.99%, 1.19%, 1.98%, and 2.36% compared to the MLP, AlexNet, and LeNet models, respectively. Moreover, the BDL-MLP model has the best performance in robustness and feature extraction abilities for complex data, and can quantify the prediction uncertainty. The strategy recognition method of combat aircraft maneuvering based on Bayesian deep learning can provide valuable insights for the research and development of military intelligent auxiliary combat decision-support systems.
- 1 July 2025
- Published by China Science Publishing & Media Ltd. in JOURNAL OF SHENZHEN UNIVERSITY SCIENCE AND ENGINEERING
- Vol. 42 (4) , 437-446
Open AccessManeuver strategy recognition technology for enemy combat aircraft based on Bayesian deep learning
Enhancing identification of enemy combat aircraft maneuver strategies is a critical factor in improving air combat decision-making capabilities. As traditional deep learning models often show overconfidence in complex and variable combat environment, and it is difficult to evaluate the uncertainty, a combat aircraft maneuver strategy recognition method based on Bayesian deep learning is proposed. The method employs Bayesian variational inference and multivariate Gaussian distribution to construct a multi-layer perceptron-based Bayesian deep learning (BDL-MLP) probabilistic model. A gradient balancing factor is introduced to reduce the imbalance between complexity cost gradient and likelihood cost gradient, and then Bayes backpropagation algorithm is used for model training and parameter optimization. Based on the virtual air combat simulation platform AirFlightSim developed in Unity3D software, BDL-MLP, multilayer perceptron (MLP), AlexNet and LeNet models are used to classify and evaluate the data sets of combat aircraft motion scenes with varying degrees of blur (blur radii of 0, 15, 31, 45, and 61 pixels, respectively). Results show that, on the five datasets constructed with the aforementioned blur radii, the maneuver strategy identification accuracy of the BDL-MLP model showed average improvements of 0.43%, 0.99%, 1.19%, 1.98%, and 2.36% compared to the MLP, AlexNet, and LeNet models, respectively. Moreover, the BDL-MLP model has the best performance in robustness and feature extraction abilities for complex data, and can quantify the prediction uncertainty. The strategy recognition method of combat aircraft maneuvering based on Bayesian deep learning can provide valuable insights for the research and development of military intelligent auxiliary combat decision-support systems.
- Open Access
- 27 June 2025
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 11 (8) , 1-17
An air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
- 27 June 2025
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 11 (8) , 1-17
Open AccessAn air combat maneuver decision-making approach using coupled reward in deep reinforcement learning
- Open Access
- 22 July 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (6) , 7513-7530
Predictive air combat decision model with segmented reward allocation
In air combat missions, unmanned combat aerial vehicles (UCAVs) must take strategic actions to establish combat advantages, enabling effective tracking and attacking of enemy UCAVs. Currently, a lot of reinforcement learning algorithms are applied to the air combat mission of unmanned fighter aircraft. However, most algorithms can only select policies based on the current state of both sides. This leads to the inability to effectively track and attack when the enemy performs large angle maneuvering. Additionally, these algorithms cannot adapt to different situations, resulting in the unmanned fighter aircraft being at a disadvantage in some cases. To solve these problems, this paper proposes predictive air combat decision model with segmented reward allocation for air combat tracking and attacking. On the basis of the air combat environment, we propose the prediction soft actor-critic (Pre-SAC) algorithm, which combines the prediction of enemy states with the states of UCAV for model training. This enables the UCAV to predict the next move of the enemy UCAV in advance and establish a greater air combat advantage for us. Furthermore, by adopting a segmented reward allocation model and combining it with the Pre-SAC algorithm, we propose the segmented reward allocation soft actor-critic (Sra-SAC) algorithm, which solves the problem of UCAVs being unable to adapt to different situations. The results show that the prediction-based segmented reward allocation the Sra-SAC algorithm outperforms the traditional soft actor-critic (SAC) algorithm in terms of overall reward, travel distance, and relative advantage.
- 22 July 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (6) , 7513-7530
Open AccessPredictive air combat decision model with segmented reward allocation
In air combat missions, unmanned combat aerial vehicles (UCAVs) must take strategic actions to establish combat advantages, enabling effective tracking and attacking of enemy UCAVs. Currently, a lot of reinforcement learning algorithms are applied to the air combat mission of unmanned fighter aircraft. However, most algorithms can only select policies based on the current state of both sides. This leads to the inability to effectively track and attack when the enemy performs large angle maneuvering. Additionally, these algorithms cannot adapt to different situations, resulting in the unmanned fighter aircraft being at a disadvantage in some cases. To solve these problems, this paper proposes predictive air combat decision model with segmented reward allocation for air combat tracking and attacking. On the basis of the air combat environment, we propose the prediction soft actor-critic (Pre-SAC) algorithm, which combines the prediction of enemy states with the states of UCAV for model training. This enables the UCAV to predict the next move of the enemy UCAV in advance and establish a greater air combat advantage for us. Furthermore, by adopting a segmented reward allocation model and combining it with the Pre-SAC algorithm, we propose the segmented reward allocation soft actor-critic (Sra-SAC) algorithm, which solves the problem of UCAVs being unable to adapt to different situations. The results show that the prediction-based segmented reward allocation the Sra-SAC algorithm outperforms the traditional soft actor-critic (SAC) algorithm in terms of overall reward, travel distance, and relative advantage.