I am an Assistant Professor at Robot Dynamics Section as part of the Department of Cognitive Robotics, Faculty of Mechanical, Maritime and Materials Engineering (3mE), Delft University of Technology.

My research goal is to establish a machine learning control framework by unifying machine learning and control theory, with applications in robotics. In particular, I am interested in Bayesian learning for system identification, (multi-agent) reinforcement learning for decision&control and state estimation.

Recent News

• October 14, 2020, Our paper “Social-VRNN: One-Shot Multi-modal Trajectory Prediction for Interacting Pedestrians” accepted to CoRL!
• September 11, 2020, Our paper “Zero-Sum Game Based Optimal Secure Control under Actuator Attacks” accepted to IEEE Transactions on Automatic Control!
• September 2, 2020, Our paper “Dynamic Network Reconstruction from Heterogeneous Datasets” accepted to Automatica!
• July 16, 2020, Our paper “Model-Reference Reinforcement Learning Control of Autonomous Surface Vehicles with Uncertainties” accepted to CDC! Check extended version
• July 7, 2020, Our paper “Lyapunov-Based Reinforcement Learning for Decentralized Multi-Agent Control” accepted to DAI！
• July 6, 2020, I will serve as an Aera Chair for CoRL 2020.
• July 1, 2020, Our paper “Actor-Critic Reinforcement Learning for Control with Stability Guarantee” accepted to RA-L and IROS!
• April 1, 2020, Dr Chengchao Bai from Harbin Institue of Technology joined our group as a PostDoc. Welcome, Chengchao!
• Febrary 3, 2020, I will serve as an Associate Editor for IROS 2020.
• Janurary 15, 2020, Two Postdoc positions openning on “Safe Reinforcement Learning for Motion Planning” funded by AnKobot. One position is on Simulator. One position is on Theory.
• Janurary 14, 2020, Our paper “Towards Lossless Binary Convolutional Neural Networks Using Piecewise Approximation” accepted to ECAI, 2020. Well done, Baozhou.
• October 25, 2019, Our paper “Data Driven Discovery of Cyber Physical Systems” is published in Nature Communications!
• October 22, 2019, Our paper “Asynchronous Observer Design for Switched Linear Systems: A Tube-Based Approach” is published in IEEE/CAA Journal of Automatica Sinica.
• October 2, 2019, Our paper “H∞ Model-free Reinforcement Learning with Robust Stability Guarantee” is accepted to NeurIPS 2019 Workshop on Robot Learning.
• September 12, 2019. Dr Qingrui Zhang from University of Toronto joined our group as a PostDoc, co-supervised by Dr Vasso Reppa from Department of Maritime and Transport Technology. Welcome, Qingrui!
• April 22, 2019. Our paper “BayesNAS: A Bayesian Approach for Neural Architecture Search” accepted to ICML 2019! Hongpeng also got the ICML Travel Award. Well done, Hongpeng and Minghao.
• April 11, 2019. Hongpeng gave a talk on “Sparse Bayesian Deep Neural Networks for Nonlinear System Identification” at Nonlinear System Identification Benchmarks, Eindhoven.
• April 1, 2019. Talk at Huawei Noah’s Ark Lab, London, UK.
• Febrary 11, 2019. One Postdoc position openning on “Learning for Self-healing of Multi-Machine Systems”, apply here
• Febrary 8, 2019. Two abstracts accepted to Benelux Meeting.
• December 21, 2018. Our paper Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning accepted to ICLR 2019!
• May 1, 2018. Start my research group in the Department of Cognitive Robotics, TU Delft. Thanks to the amazing colleagues in DJI! Thanks Shenzhen!