Zhenfang Ma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhenfang Ma has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Kaizhou Gao
Macau Institute of Systems Engineering, Macau University of Science and Technology, Taipa, Macau, China
3 shared publications - Hui Yu
Macau Institute of Systems Engineering, Macau University of Science and Technology, Macao 999078, China
2 shared publications - Ling Wang
Tsinghua University
1 shared publication Mengchu Zhou
1 shared publicationMinglong Gao
1 shared publication- Naiqi Wu
Institute of Systems Engineering, Macau University of Science and Technology, Macau, China
1 shared publication
Latest Publications
- 17 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 29 (5) , 1783-1794
Exact and Deep Q-Network Assisted Swarm Intelligence Methods for Scheduling Multiobjective Heterogeneous Unmanned Surface Vehicles
In complex navigational environments, effective unmanned surface vehicle (USV) scheduling is critical. However, the obstacle avoidance problems are often ignored in the literature. This study addresses the multi-objective heterogeneous USV scheduling problems with obstacle avoidance. The objective is to minimize the maximum completion time and total carbon emissions. First, a mathematical model is developed to describe the concerned problems. Second, an A* algorithm is employed to obtain a path between task points with avoiding obstacles. Third, to obtain a high-quality scheduling scheme, an improved artificial bee colony (ABC) algorithm with deep Q-network (DQN) is designed. According to the problem nature, six novel rules are designed for finding high-quality solutions in initialized population. Five local search strategies are designed based on the structure of solution space. The scales of the instances and different objectives are utilized to designed a DQN, which recommends suitable local search strategies during iterations for improving convergence speed. Then, the Gurobi solver is employed to verify the proposed model. The effectiveness of the proposed strategies is verified by 13 instances with different scales. Finally, the experimental results and discussions show that the designed ABC with DQN has the strongest competitiveness among all compared algorithms.
- 17 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Evolutionary Computation
- Vol. 29 (5) , 1783-1794
Exact and Deep Q-Network Assisted Swarm Intelligence Methods for Scheduling Multiobjective Heterogeneous Unmanned Surface Vehicles
In complex navigational environments, effective unmanned surface vehicle (USV) scheduling is critical. However, the obstacle avoidance problems are often ignored in the literature. This study addresses the multi-objective heterogeneous USV scheduling problems with obstacle avoidance. The objective is to minimize the maximum completion time and total carbon emissions. First, a mathematical model is developed to describe the concerned problems. Second, an A* algorithm is employed to obtain a path between task points with avoiding obstacles. Third, to obtain a high-quality scheduling scheme, an improved artificial bee colony (ABC) algorithm with deep Q-network (DQN) is designed. According to the problem nature, six novel rules are designed for finding high-quality solutions in initialized population. Five local search strategies are designed based on the structure of solution space. The scales of the instances and different objectives are utilized to designed a DQN, which recommends suitable local search strategies during iterations for improving convergence speed. Then, the Gurobi solver is employed to verify the proposed model. The effectiveness of the proposed strategies is verified by 13 instances with different scales. Finally, the experimental results and discussions show that the designed ABC with DQN has the strongest competitiveness among all compared algorithms.
- Top Cited Papers
- 21 February 2024
- Published by Elsevier in Swarm and Evolutionary Computation
- Vol. 86
Artificial intelligence algorithms in unmanned surface vessel task assignment and path planning: A survey
- 21 February 2024
- Published by Elsevier in Swarm and Evolutionary Computation
- Vol. 86
Top Cited PapersArtificial intelligence algorithms in unmanned surface vessel task assignment and path planning: A survey
- Open Access
- 19 January 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (2) , 339
Solving Heterogeneous USV Scheduling Problems by Problem-Specific Knowledge Based Meta-Heuristics with Q-Learning
This study focuses on the scheduling problem of heterogeneous unmanned surface vehicles (USVs) with obstacle avoidance pretreatment. The goal is to minimize the overall maximum completion time of USVs. First, we develop a mathematical model for the problem. Second, with obstacles, an A* algorithm is employed to generate a path between two points where tasks need to be performed. Third, three meta-heuristics, i.e., simulated annealing (SA), genetic algorithm (GA), and harmony search (HS), are employed and improved to solve the problems. Based on problem-specific knowledge, nine local search operators are designed to improve the performance of the proposed algorithms. In each iteration, three Q-learning strategies are used to select high-quality local search operators. We aim to improve the performance of meta-heuristics by using Q-learning-based local search operators. Finally, 13 instances with different scales are adopted to validate the effectiveness of the proposed strategies. We compare with the classical meta-heuristics and the existing meta-heuristics. The proposed meta-heuristics with Q-learning are overall better than the compared ones. The results and comparisons show that HS with the second Q-learning, HS + QL2, exhibits the strongest competitiveness (the smallest mean rank value 1.00) among 15 algorithms.
- 19 January 2024
- Published by MDPI AG in Mathematics
- Vol. 12 (2) , 339
Open AccessSolving Heterogeneous USV Scheduling Problems by Problem-Specific Knowledge Based Meta-Heuristics with Q-Learning
This study focuses on the scheduling problem of heterogeneous unmanned surface vehicles (USVs) with obstacle avoidance pretreatment. The goal is to minimize the overall maximum completion time of USVs. First, we develop a mathematical model for the problem. Second, with obstacles, an A* algorithm is employed to generate a path between two points where tasks need to be performed. Third, three meta-heuristics, i.e., simulated annealing (SA), genetic algorithm (GA), and harmony search (HS), are employed and improved to solve the problems. Based on problem-specific knowledge, nine local search operators are designed to improve the performance of the proposed algorithms. In each iteration, three Q-learning strategies are used to select high-quality local search operators. We aim to improve the performance of meta-heuristics by using Q-learning-based local search operators. Finally, 13 instances with different scales are adopted to validate the effectiveness of the proposed strategies. We compare with the classical meta-heuristics and the existing meta-heuristics. The proposed meta-heuristics with Q-learning are overall better than the compared ones. The results and comparisons show that HS with the second Q-learning, HS + QL2, exhibits the strongest competitiveness (the smallest mean rank value 1.00) among 15 algorithms.