Yintao Wang
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
Yintao Wang has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Qi Sun
Northwestern Polytechnical University
8 shared publications - Keyang Wang
Northwestern Polytechnical University
5 shared publications Peng Chang
3 shared publicationsQi Sun
3 shared publicationsYao Yao
Jiangsu Automation Research Institute Lianyungang China
3 shared publicationsGuanglei Song
2 shared publications
Latest Publications
- 24 June 2026
- Published by Elsevier in Ocean Engineering
- Vol. 363
Research on three-dimensional path planning algorithms for safe and covert underwater navigation
- 24 June 2026
- Published by Elsevier in Ocean Engineering
- Vol. 363
Research on three-dimensional path planning algorithms for safe and covert underwater navigation
- 7 June 2026
- Published by Elsevier in Ocean Engineering
- Vol. 362
ReGA-PINN: A physics-informed neural network approach for dynamics modeling and control of remotely operated vehicles
- 7 June 2026
- Published by Elsevier in Ocean Engineering
- Vol. 362
ReGA-PINN: A physics-informed neural network approach for dynamics modeling and control of remotely operated vehicles
- Open Access
- 23 April 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (9) , 775
Search-Information-Driven Collaborative Task Planning for Multi-UUV Systems
To address the problems of unreasonable task allocation and low target search efficiency in the collaborative search of multiple unmanned undersea vehicles (UUVs) in complex marine environments, this paper proposes a search-information-driven collaborative task planning method for multi-UUV systems, and constructs a systematic and integrated multi-UUV collaborative task planning framework. Considering the spatial characteristics of the complex underwater environment and sonar detection rules, an underwater task environment grid model and an active sonar instantaneous detection model are constructed as the environmental and detection foundation of the framework. Within the framework, the Gaussian Mixture Model (GMM) is adopted to realize dynamic division of task regions, and reasonable resource allocation among multiple UUVs is achieved by defining scientific area allocation indicators. A search information map consisting of target probability distribution and environmental uncertainty is established, and a receding horizon planning framework is introduced to balance short-term detection effectiveness and long-term search value. Furthermore, a motion-coded Grey Wolf Optimization (GWO) algorithm is proposed to generate continuous UUV paths, which avoids path discontinuity caused by discrete grids and ensures the convergence efficiency of the algorithm. Simulation results verify that compared with traditional methods, the proposed method improves the total probability benefit by 19.87% and the number of discovered targets by 18.29%, demonstrating better search performance and environmental adaptability.
- 23 April 2026
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 14 (9) , 775
Open AccessSearch-Information-Driven Collaborative Task Planning for Multi-UUV Systems
To address the problems of unreasonable task allocation and low target search efficiency in the collaborative search of multiple unmanned undersea vehicles (UUVs) in complex marine environments, this paper proposes a search-information-driven collaborative task planning method for multi-UUV systems, and constructs a systematic and integrated multi-UUV collaborative task planning framework. Considering the spatial characteristics of the complex underwater environment and sonar detection rules, an underwater task environment grid model and an active sonar instantaneous detection model are constructed as the environmental and detection foundation of the framework. Within the framework, the Gaussian Mixture Model (GMM) is adopted to realize dynamic division of task regions, and reasonable resource allocation among multiple UUVs is achieved by defining scientific area allocation indicators. A search information map consisting of target probability distribution and environmental uncertainty is established, and a receding horizon planning framework is introduced to balance short-term detection effectiveness and long-term search value. Furthermore, a motion-coded Grey Wolf Optimization (GWO) algorithm is proposed to generate continuous UUV paths, which avoids path discontinuity caused by discrete grids and ensures the convergence efficiency of the algorithm. Simulation results verify that compared with traditional methods, the proposed method improves the total probability benefit by 19.87% and the number of discovered targets by 18.29%, demonstrating better search performance and environmental adaptability.
- 15 July 2025
- Published by Elsevier in Systems & Control Letters
- Vol. 204
A time advantage function-based solution to two-player circular target defense games
- 15 July 2025
- Published by Elsevier in Systems & Control Letters
- Vol. 204
A time advantage function-based solution to two-player circular target defense games
- 11 July 2025
- Published by World Scientific Pub Co Pte Ltd in Unmanned Systems
- p. 1-20
A Research on Noncontact Confrontation for Target Defense Against a Superior Intruder
This paper investigates a target-defense (TD) scenario involving inferior defenders protecting a restricted area against a superior intruder. The intruder’s high-speed mobility precludes traditional capture strategies, necessitating a novel noncontact confrontation paradigm. In this framework, we introduce a threat radius determined by the defender’s weapon range as a critical operational parameter. The target region imposes movement constraints on defenders, while the threat radius fundamentally alters the engagement dynamics, rendering classical differential game methods inapplicable. To address these challenges, we develop an integrated approach combining geometric methods and theoretical analysis. Leveraging the noncontact confrontation concept, we design innovative interception strategies for both single and multiple defender configurations. This noncontact paradigm offers a new perspective and effective methodology for addressing complex TD problems.
- 11 July 2025
- Published by World Scientific Pub Co Pte Ltd in Unmanned Systems
- p. 1-20
A Research on Noncontact Confrontation for Target Defense Against a Superior Intruder
This paper investigates a target-defense (TD) scenario involving inferior defenders protecting a restricted area against a superior intruder. The intruder’s high-speed mobility precludes traditional capture strategies, necessitating a novel noncontact confrontation paradigm. In this framework, we introduce a threat radius determined by the defender’s weapon range as a critical operational parameter. The target region imposes movement constraints on defenders, while the threat radius fundamentally alters the engagement dynamics, rendering classical differential game methods inapplicable. To address these challenges, we develop an integrated approach combining geometric methods and theoretical analysis. Leveraging the noncontact confrontation concept, we design innovative interception strategies for both single and multiple defender configurations. This noncontact paradigm offers a new perspective and effective methodology for addressing complex TD problems.
- Open Access
- 16 February 2025
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 13 (2) , 365
Target-Defense Games with One or Two Unmanned Surface Vehicles Defending an Island: A Geometric Analytical Approach
This paper considers a target-defense game in an open area with one or two defenders as well as an intruder. The intruder endeavors to reach the boundary of the island, while the defenders strive to prevent that by capturing the intruder through contact. Islands, as closed areas, restrict the free movement of the defenders, since the defenders—represented by USVs—cannot traverse the target area directly. First, we are concerned with the barrier, which is the boundary of the winning zones, taking into account the impact of the target. For the initial states lying in the defenders’ winning zone, there exists a strategy for the defenders to intercept the intruder regardless of the intruder’s best effort, while for the initial states lying in the intruder’s winning zone, the intruder can always invade successfully. We propose a geometric method to construct the barrier analytically for two kinds of speed ratios. Then, by taking index functions into consideration, we present optimal strategies for the players after constructing the dominance regions when their initial states lie in different winning zones. Simulation results verify the effectiveness of the proposed method. This study can be extended to scenarios involving multiple defenders.
- 16 February 2025
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 13 (2) , 365
Open AccessTarget-Defense Games with One or Two Unmanned Surface Vehicles Defending an Island: A Geometric Analytical Approach
This paper considers a target-defense game in an open area with one or two defenders as well as an intruder. The intruder endeavors to reach the boundary of the island, while the defenders strive to prevent that by capturing the intruder through contact. Islands, as closed areas, restrict the free movement of the defenders, since the defenders—represented by USVs—cannot traverse the target area directly. First, we are concerned with the barrier, which is the boundary of the winning zones, taking into account the impact of the target. For the initial states lying in the defenders’ winning zone, there exists a strategy for the defenders to intercept the intruder regardless of the intruder’s best effort, while for the initial states lying in the intruder’s winning zone, the intruder can always invade successfully. We propose a geometric method to construct the barrier analytically for two kinds of speed ratios. Then, by taking index functions into consideration, we present optimal strategies for the players after constructing the dominance regions when their initial states lie in different winning zones. Simulation results verify the effectiveness of the proposed method. This study can be extended to scenarios involving multiple defenders.
- 25 January 2025
- Published by Springer Nature
- p. 80-93
Informative Path Planning for Multi-UUV Cooperative Search with Distributed Q-Learning
- 25 January 2025
- Published by Springer Nature
- p. 80-93
Informative Path Planning for Multi-UUV Cooperative Search with Distributed Q-Learning