Wenhao Tan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wenhao Tan has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Ran Song
School of Control Science and Engineering, Shandong University, Jinan, China
7 shared publications - Wei Zhang
Shandong University
7 shared publications - Xing Fang
School of Control Science and Engineering, Shandong University, Jinan, China
4 shared publications - Yanyun Chen
Shandong University
4 shared publications - Zhiheng Li
School of Control Science and Engineering, Shandong University, Jinan, China
4 shared publications - Qian Zhang
Shandong University
2 shared publications
Latest Publications
- Open Access
HAML: Humanoid Adversarial Multi-Skill Learning via a Single Policy
Translating large-scale motion datasets into robust, deployable humanoid controllers is a critical challenge in engineering informatics, primarily due to the scarcity of high-quality annotations, the risk of mode collapse in conditional generation, and the strict constraints of onboard computing hardware. This paper presents a deployable two-stage learning system that maps clip-level motion datasets to a single-policy multi-skill controller and its deployable counterpart. We adopt coarse one-hot skill labels that can be assigned automatically at the clip level with negligible manual effort, enabling scalable dataset construction. To prevent conditional discriminators from ignoring skill conditions, we inject mismatched (transition, label) pairs and introduce a condition-aware loss that explicitly penalizes incorrect transition–label associations, improving controllability and mitigating mode collapse. For real-world deployment, we further propose a two-stage training strategy: a privileged teacher policy is first trained in simulation and then distilled into a student policy that relies on stacked historical proprioceptive observations, ensuring robustness against sensing noise and latency without relying on external state estimation. Extensive evaluations in simulation and on real hardware demonstrate improved skill coverage, transition coverage, realism, and training efficiency across heterogeneous embodiments. With the onboard computer of a Unitree G1 robot, the distilled policy runs at 100 Hz with 15–25 ms latency, confirming the system’s engineering feasibility.
Open AccessHAML: Humanoid Adversarial Multi-Skill Learning via a Single Policy
Translating large-scale motion datasets into robust, deployable humanoid controllers is a critical challenge in engineering informatics, primarily due to the scarcity of high-quality annotations, the risk of mode collapse in conditional generation, and the strict constraints of onboard computing hardware. This paper presents a deployable two-stage learning system that maps clip-level motion datasets to a single-policy multi-skill controller and its deployable counterpart. We adopt coarse one-hot skill labels that can be assigned automatically at the clip level with negligible manual effort, enabling scalable dataset construction. To prevent conditional discriminators from ignoring skill conditions, we inject mismatched (transition, label) pairs and introduce a condition-aware loss that explicitly penalizes incorrect transition–label associations, improving controllability and mitigating mode collapse. For real-world deployment, we further propose a two-stage training strategy: a privileged teacher policy is first trained in simulation and then distilled into a student policy that relies on stacked historical proprioceptive observations, ensuring robustness against sensing noise and latency without relying on external state estimation. Extensive evaluations in simulation and on real hardware demonstrate improved skill coverage, transition coverage, realism, and training efficiency across heterogeneous embodiments. With the onboard computer of a Unitree G1 robot, the distilled policy runs at 100 Hz with 15–25 ms latency, confirming the system’s engineering feasibility.
- 23 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (8) , 11949-11960
MI-MRL: A Meta-Reinforcement Learning Method Based on Memory Information for Bipedal Locomotion Task
Bipedal locomotion is challenging for many learning-based algorithms. The sim-to-real gap, the changes in task loads, and the changes in joint damping caused by long-term use may all degrade the performance of these algorithms. Aiming at such issues, we propose a memory-informed meta-reinforcement learning (MI-MRL) method for bipedal locomotion. MI-MRL demonstrates robustness to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment. It replaces the standard feedback rewards commonly used in MRL and restructures the input to the context encoder, thus broadening the applicability of MRL-based methods to real-world scenarios. MI-MRL also introduces a state estimator and a noise filter to accelerate training and enhance algorithm performance. According to the experimental results, MI-MRL completes information collection and context adaptation within 0.32 s, and achieves an external load tolerance 20 times higher than that of the latest MRL-based methods. The results also demonstrate that MI-MRL is robust to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment.
- 23 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (8) , 11949-11960
MI-MRL: A Meta-Reinforcement Learning Method Based on Memory Information for Bipedal Locomotion Task
Bipedal locomotion is challenging for many learning-based algorithms. The sim-to-real gap, the changes in task loads, and the changes in joint damping caused by long-term use may all degrade the performance of these algorithms. Aiming at such issues, we propose a memory-informed meta-reinforcement learning (MI-MRL) method for bipedal locomotion. MI-MRL demonstrates robustness to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment. It replaces the standard feedback rewards commonly used in MRL and restructures the input to the context encoder, thus broadening the applicability of MRL-based methods to real-world scenarios. MI-MRL also introduces a state estimator and a noise filter to accelerate training and enhance algorithm performance. According to the experimental results, MI-MRL completes information collection and context adaptation within 0.32 s, and achieves an external load tolerance 20 times higher than that of the latest MRL-based methods. The results also demonstrate that MI-MRL is robust to various task-related parameter changes and offers a practical solution to the difficulties that MRL faces in real-world deployment.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23, 11868-11880
MoTL: Modality-Balanced Terrain-Aware Locomotion Learning for Quadruped Robots
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23, 11868-11880
MoTL: Modality-Balanced Terrain-Aware Locomotion Learning for Quadruped Robots
- 9 November 2025
- Published by Springer Nature
- p. 281-293
Learning Smooth and Coordinated Quadruped Motions via Incremental Foot Position Control
- 9 November 2025
- Published by Springer Nature
- p. 281-293
Learning Smooth and Coordinated Quadruped Motions via Incremental Foot Position Control
- 24 October 2025
- Published by Springer Nature
- p. 662-672
Cluster-Guided State Initialization Strategy for Flexible Humanoid Locomotion
- 24 October 2025
- Published by Springer Nature
- p. 662-672
Cluster-Guided State Initialization Strategy for Flexible Humanoid Locomotion
- 24 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (9) , 8138-8145
MCLER: Multi-Critic Continual Learning With Experience Replay for Quadruped Gait Generation
Quadruped robots are able to traverse most terrains on the earth using a wide variety of gaits, providing solutions for robots to operate in specialized environments. Although existing methods have achieved excellent performance in gait generation, they suffer from catastrophic forgetting and inability to evolve when encountering new gait demands. Fortunately, continual learning provides a powerful tool to address this issue and has demonstrated the effectiveness in classification, recognition, and other fields but not gait generation. This paper presents a framework of multi-critic continual learning with experience replay (MCLER) to enable incremental gait learning. MCLER allows quadruped robots to be lifelong learners and acquire new gaits without forgetting. Extensive experiments conducted in simulated and real environments show that MCLER significantly outperforms the state-of-the-art methods.
- 24 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (9) , 8138-8145
MCLER: Multi-Critic Continual Learning With Experience Replay for Quadruped Gait Generation
Quadruped robots are able to traverse most terrains on the earth using a wide variety of gaits, providing solutions for robots to operate in specialized environments. Although existing methods have achieved excellent performance in gait generation, they suffer from catastrophic forgetting and inability to evolve when encountering new gait demands. Fortunately, continual learning provides a powerful tool to address this issue and has demonstrated the effectiveness in classification, recognition, and other fields but not gait generation. This paper presents a framework of multi-critic continual learning with experience replay (MCLER) to enable incremental gait learning. MCLER allows quadruped robots to be lifelong learners and acquire new gaits without forgetting. Extensive experiments conducted in simulated and real environments show that MCLER significantly outperforms the state-of-the-art methods.
- 23 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 46 (12) , 8396-8409
IGCN: A Provably Informative GCN Embedding for Semi-Supervised Learning With Extremely Limited Labels
Graph Neural Networks (GNNs) have gained much more attention in the representation learning for the graph-structured data. However, the labels are always limited in the graph, which easily leads to the overfitting problem and causes the poor performance. To solve this problem, we propose a new framework called IGCN, short for Informative Graph Convolutional Network, where the objective of IGCN is designed to obtain the informative embeddings via discarding the task-irrelevant information of the graph data based on the mutual information. As the mutual information for irregular data is intractable to compute, our framework is optimized via a surrogate objective, where two terms are derived to approximate the original objective. For the former term, it demonstrates that the mutual information between the learned embeddings and the ground truth should be high, where we utilize the semi-supervised classification loss and the prototype based supervised contrastive learning loss for optimizing it. For the latter term, it requires that the mutual information between the learned node embeddings and the initial embeddings should be high and we propose to minimize the reconstruction loss between them to achieve the goal of maximizing the latter term from the feature level and the layer level, which contains the graph encoder-decoder module and a novel architecture GCN$_{Info}$ . Moreover, we provably show that the designed GCN$_{Info}$ can better alleviate the information loss and preserve as much useful information of the initial embeddings as possible. Experimental results show that the IGCN outperforms the state-of-the-art methods on 7 popular datasets.- 23 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 46 (12) , 8396-8409
IGCN: A Provably Informative GCN Embedding for Semi-Supervised Learning With Extremely Limited Labels
Graph Neural Networks (GNNs) have gained much more attention in the representation learning for the graph-structured data. However, the labels are always limited in the graph, which easily leads to the overfitting problem and causes the poor performance. To solve this problem, we propose a new framework called IGCN, short for Informative Graph Convolutional Network, where the objective of IGCN is designed to obtain the informative embeddings via discarding the task-irrelevant information of the graph data based on the mutual information. As the mutual information for irregular data is intractable to compute, our framework is optimized via a surrogate objective, where two terms are derived to approximate the original objective. For the former term, it demonstrates that the mutual information between the learned embeddings and the ground truth should be high, where we utilize the semi-supervised classification loss and the prototype based supervised contrastive learning loss for optimizing it. For the latter term, it requires that the mutual information between the learned node embeddings and the initial embeddings should be high and we propose to minimize the reconstruction loss between them to achieve the goal of maximizing the latter term from the feature level and the layer level, which contains the graph encoder-decoder module and a novel architecture GCN$_{Info}$ . Moreover, we provably show that the designed GCN$_{Info}$ can better alleviate the information loss and preserve as much useful information of the initial embeddings as possible. Experimental results show that the IGCN outperforms the state-of-the-art methods on 7 popular datasets.- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 21 (4) , 5367-5378
A Hierarchical Framework for Quadruped Omnidirectional Locomotion Based on Reinforcement Learning
Quadruped locomotion is challenging for many learning-based algorithms. This is because it requires tedious manual tuning to cope with different types of terrains and is difficult to deploy in reality due to the sim-to-real gap between the training and the testing scenarios. This paper proposes a quadruped robot learning system for agile locomotion which does not require any pre-training and works well in various terrains. We introduce a hierarchical framework that uses reinforcement learning as the high-level policy to adjust the low-level trajectory generator for a better adaptability to various terrains. We compact the observation and the action spaces of reinforcement learning to deploy the proposed framework on a host computer interfaced with the robot. Besides, we design an omnidirectional trajectory generator guided by robot posture, which generates omnidirectional foot trajectories to interact with the environment. Experimental results and the supplementary video demonstrate that our hierarchical framework only trained in simulation can be easily deployed in the real world, and also has the advantages of fast convergence and good terrain adaptability. Note to Practitioners —This paper presents a hierarchical framework for quadruped robots. It combines a high-level reinforcement learning controller with a posture-guided trajectory generator to adaptively generate omnidirectional motions. Our method is easy to train as it converges fast and does not need to adjust a dozen or so of rewards. The quadruped robot can be deployed in a real environment directly after being trained in simulation. With the trained hierarchical framework deployed on a remote host computer, the robot works well in a variety of real-world environments unseen in the simulation.
- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 21 (4) , 5367-5378
A Hierarchical Framework for Quadruped Omnidirectional Locomotion Based on Reinforcement Learning
Quadruped locomotion is challenging for many learning-based algorithms. This is because it requires tedious manual tuning to cope with different types of terrains and is difficult to deploy in reality due to the sim-to-real gap between the training and the testing scenarios. This paper proposes a quadruped robot learning system for agile locomotion which does not require any pre-training and works well in various terrains. We introduce a hierarchical framework that uses reinforcement learning as the high-level policy to adjust the low-level trajectory generator for a better adaptability to various terrains. We compact the observation and the action spaces of reinforcement learning to deploy the proposed framework on a host computer interfaced with the robot. Besides, we design an omnidirectional trajectory generator guided by robot posture, which generates omnidirectional foot trajectories to interact with the environment. Experimental results and the supplementary video demonstrate that our hierarchical framework only trained in simulation can be easily deployed in the real world, and also has the advantages of fast convergence and good terrain adaptability. Note to Practitioners —This paper presents a hierarchical framework for quadruped robots. It combines a high-level reinforcement learning controller with a posture-guided trajectory generator to adaptively generate omnidirectional motions. Our method is easy to train as it converges fast and does not need to adjust a dozen or so of rewards. The quadruped robot can be deployed in a real environment directly after being trained in simulation. With the trained hierarchical framework deployed on a remote host computer, the robot works well in a variety of real-world environments unseen in the simulation.