Baofu Fang
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
Baofu Fang has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
Hao Wang
Hefei University of Technology
7 shared publications- Xiaohui Yuan
University of North Texas
3 shared publications - Caiming Zheng
School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China
2 shared publications - Hao Wang
School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China
2 shared publications - Qiubo Zhong
Ningbo University of Technology
2 shared publications Shuanglu Dai
School of Computer and Information, Hefei University of Technology, Hefei, China
2 shared publications
Latest Publications
- Open Access
- 27 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 17 (3) , 1-19
PMARL: Multi-Agent Reinforcement Learning in Large-Scale Systems
Large-scale multi-agent systems face two core challenges: inefficient policy learning and the explosion of state dimensions. Existing methods often rely on manually designed task sequences to guide agents’ learning in stages, but these designs lack adaptability to agents’ learning abilities, making it difficult to ensure the rationality of task difficulty. Moreover, the representation capability of current network structures is limited, making it challenging to efficiently handle high-dimensional state information and complex interaction relationships. To address these issues, we propose a Progressive Multi-Agent Reinforcement Learning (PMARL) framework. PMARL introduces a task adapter that adaptively selects task difficulty based on agents’ learning abilities, eliminating reliance on manual experience. Additionally, a Dynamic Dimension Adaptive Network (DDAN) is designed, incorporating hypernetwork and self-attention mechanisms to achieve adaptive feature extraction of high-dimensional states and efficient representation of agent interaction relationships. Experimental results demonstrate that PMARL exhibits higher efficiency and better adaptability compared to existing methods when addressing large-scale multi-agent tasks.
- 27 April 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 17 (3) , 1-19
Open AccessPMARL: Multi-Agent Reinforcement Learning in Large-Scale Systems
Large-scale multi-agent systems face two core challenges: inefficient policy learning and the explosion of state dimensions. Existing methods often rely on manually designed task sequences to guide agents’ learning in stages, but these designs lack adaptability to agents’ learning abilities, making it difficult to ensure the rationality of task difficulty. Moreover, the representation capability of current network structures is limited, making it challenging to efficiently handle high-dimensional state information and complex interaction relationships. To address these issues, we propose a Progressive Multi-Agent Reinforcement Learning (PMARL) framework. PMARL introduces a task adapter that adaptively selects task difficulty based on agents’ learning abilities, eliminating reliance on manual experience. Additionally, a Dynamic Dimension Adaptive Network (DDAN) is designed, incorporating hypernetwork and self-attention mechanisms to achieve adaptive feature extraction of high-dimensional states and efficient representation of agent interaction relationships. Experimental results demonstrate that PMARL exhibits higher efficiency and better adaptability compared to existing methods when addressing large-scale multi-agent tasks.
- 12 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Fuzzy Systems
- Vol. 31 (2) , 511-520
Two-Stream Fused Fuzzy Deep Neural Network for Multiagent Learning
In multi-agent reinforcement learning, multi-layer fully connected neural network(MLP) is used for value function approximation, which solves large-scale or continuous space problems. However, it is easy fall into a local optimal and overfitting under partially observed environments. Because each agent lacks the information that plays a key role in decision-making beyond the observation field. Even if communication is allowed, the received informations in communication channel have large noise due to the observations of other agents and strong uncertainty if the agnet's policy is used as the communication information. To tackle this problem, two-stream fused fuzzy deep neural network(2s-FDNN) was proposed to reduce the uncertainty and noise of information in communication channel. It is a parallel structure in which fuzzy inference module reduce the uncertainty of information and deep neural module reduce the noise of information. Then, we presented Fuzzy MA2C which integrate 2s-FDNN into multi-agent deep reinforcement learning to deal with uncertain communication informations for improving the robustness and generalization under partially observed environments. We empirically evaluate our methods in two large-scale traffic signal control environments using SUMO simulator. Results demonstrate that our methods can achieve superior performance against existing reinforcement learning algorithms.
- 12 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Fuzzy Systems
- Vol. 31 (2) , 511-520
Two-Stream Fused Fuzzy Deep Neural Network for Multiagent Learning
In multi-agent reinforcement learning, multi-layer fully connected neural network(MLP) is used for value function approximation, which solves large-scale or continuous space problems. However, it is easy fall into a local optimal and overfitting under partially observed environments. Because each agent lacks the information that plays a key role in decision-making beyond the observation field. Even if communication is allowed, the received informations in communication channel have large noise due to the observations of other agents and strong uncertainty if the agnet's policy is used as the communication information. To tackle this problem, two-stream fused fuzzy deep neural network(2s-FDNN) was proposed to reduce the uncertainty and noise of information in communication channel. It is a parallel structure in which fuzzy inference module reduce the uncertainty of information and deep neural module reduce the noise of information. Then, we presented Fuzzy MA2C which integrate 2s-FDNN into multi-agent deep reinforcement learning to deal with uncertain communication informations for improving the robustness and generalization under partially observed environments. We empirically evaluate our methods in two large-scale traffic signal control environments using SUMO simulator. Results demonstrate that our methods can achieve superior performance against existing reinforcement learning algorithms.
- Open Access
Target Localization in Wireless Sensor Networks Based on Received Signal Strength and Convex Relaxation
A new positioning algorithm based on RSS measurement is proposed. The algorithm adopts maximum likelihood estimation and semi-definite programming. The received signal strength model is transformed to a non-convex estimator for the positioning of the target using the maximum likelihood estimation. The non-convex estimator is then transformed into a convex estimator by semi-definite programming, and the global minimum of the target location estimation is obtained. This algorithm aims at the known problem and then extends its application to the case of unknown. The simulations and experimental results show that the proposed algorithm has better accuracy than the existing positioning algorithms.Open AccessTarget Localization in Wireless Sensor Networks Based on Received Signal Strength and Convex Relaxation
A new positioning algorithm based on RSS measurement is proposed. The algorithm adopts maximum likelihood estimation and semi-definite programming. The received signal strength model is transformed to a non-convex estimator for the positioning of the target using the maximum likelihood estimation. The non-convex estimator is then transformed into a convex estimator by semi-definite programming, and the global minimum of the target location estimation is obtained. This algorithm aims at the known problem and then extends its application to the case of unknown. The simulations and experimental results show that the proposed algorithm has better accuracy than the existing positioning algorithms. - Open Access
- 26 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 80028-80042
Task Allocation for Affective Robots Based on Willingness
Affection is regarded as a new solution to coordinate task allocation problems in the multi-robot system. In this paper, we proposed an interpretable and computational affection model based on willingness and a new personality model named CASE to measure the heterogeneity of the robot. Emotion contagion model is used to calculate affective interaction among robots, and we use willingness to quantify the impact of affective factors on task allocation and as an important factor for task allocation. Task allocation algorithm is designed for affective robots, and simulate the application of the affection model at task allocation by a typical example. It displays the characteristics and novelty of affection in task allocation and to verify the correctness, effectiveness, and novelty of affective robot task allocation algorithm. The result of the simulation experiment shows that affective robots in multi-robot cooperation systems obtain a better solution than rational robots and express the state of the art of task allocation for effective robots.
- 26 May 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 80028-80042
Open AccessTask Allocation for Affective Robots Based on Willingness
Affection is regarded as a new solution to coordinate task allocation problems in the multi-robot system. In this paper, we proposed an interpretable and computational affection model based on willingness and a new personality model named CASE to measure the heterogeneity of the robot. Emotion contagion model is used to calculate affective interaction among robots, and we use willingness to quantify the impact of affective factors on task allocation and as an important factor for task allocation. Task allocation algorithm is designed for affective robots, and simulate the application of the affection model at task allocation by a typical example. It displays the characteristics and novelty of affection in task allocation and to verify the correctness, effectiveness, and novelty of affective robot task allocation algorithm. The result of the simulation experiment shows that affective robots in multi-robot cooperation systems obtain a better solution than rational robots and express the state of the art of task allocation for effective robots.
- 5 May 2021
- Published by Springer Nature
- p. 594-601
Robust Visual Odometry Using Semantic Information in Complex Dynamic Scenes
- 5 May 2021
- Published by Springer Nature
- p. 594-601
Robust Visual Odometry Using Semantic Information in Complex Dynamic Scenes
- 5 May 2021
- Published by Springer Nature
- p. 309-316
Large-Scale Multi-agent Reinforcement Learning Based on Weighted Mean Field
- 5 May 2021
- Published by Springer Nature
- p. 309-316
Large-Scale Multi-agent Reinforcement Learning Based on Weighted Mean Field
- 1 January 2021
- Published by Springer Nature
- p. 334-344
RGB-D Visual Odometry Based on Semantic Feature Points in Dynamic Environments
- 1 January 2021
- Published by Springer Nature
- p. 334-344
RGB-D Visual Odometry Based on Semantic Feature Points in Dynamic Environments
- Open Access
- 1 March 2020
- Published by SAGE Publications in International Journal of Advanced Robotic Systems
- Vol. 17 (2)
A visual SLAM method based on point-line fusion in weak-matching scene
Visual simultaneous localization and mapping (SLAM) is well-known to be one of the research areas in robotics. There are many challenges in traditional point feature-based approaches, such as insufficient point features, motion jitter, and low localization accuracy in low-texture scenes, which reduce the performance of the algorithms. In this article, we propose an RGB-D SLAM system to handle these situations, which is named Point-Line Fusion (PLF)-SLAM. We utilize both points and line segments throughout the process of our work. Specifically, we present a new line segment extraction method to solve the overlap or branch problem of the line segments, and then a more rigorous screening mechanism is proposed in the line matching section. Instead of minimizing the reprojection error of points, we introduce the reprojection error based on points and lines to get a more accurate tracking pose. In addition, we come up with a solution to handle the jitter frame, which greatly improves tracking success rate and availability of the system. We thoroughly evaluate our system on the Technische Universität München (TUM) RGB-D benchmark and compare it with ORB-SLAM2, presumably the current state-of-the-art solution. The experiments show that our system has better accuracy and robustness compared to the ORB-SLAM2.
- 1 March 2020
- Published by SAGE Publications in International Journal of Advanced Robotic Systems
- Vol. 17 (2)
Open AccessA visual SLAM method based on point-line fusion in weak-matching scene
Visual simultaneous localization and mapping (SLAM) is well-known to be one of the research areas in robotics. There are many challenges in traditional point feature-based approaches, such as insufficient point features, motion jitter, and low localization accuracy in low-texture scenes, which reduce the performance of the algorithms. In this article, we propose an RGB-D SLAM system to handle these situations, which is named Point-Line Fusion (PLF)-SLAM. We utilize both points and line segments throughout the process of our work. Specifically, we present a new line segment extraction method to solve the overlap or branch problem of the line segments, and then a more rigorous screening mechanism is proposed in the line matching section. Instead of minimizing the reprojection error of points, we introduce the reprojection error based on points and lines to get a more accurate tracking pose. In addition, we come up with a solution to handle the jitter frame, which greatly improves tracking success rate and availability of the system. We thoroughly evaluate our system on the Technische Universität München (TUM) RGB-D benchmark and compare it with ORB-SLAM2, presumably the current state-of-the-art solution. The experiments show that our system has better accuracy and robustness compared to the ORB-SLAM2.
- Open Access
- 28 January 2020
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 76 (2) , 755
Advanced parallel and distributed computing for big urban data
- 28 January 2020
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 76 (2) , 755
Open AccessAdvanced parallel and distributed computing for big urban data