李浩然 男 硕导 中国科学院自动化研究所
电子邮件： lihaoran2015@ia.ac.cn
通信地址： 北京市中关村东路95号智能化大厦
邮政编码： 100190
电子邮件： lihaoran2015@ia.ac.cn
通信地址： 北京市中关村东路95号智能化大厦
邮政编码： 100190
研究领域
主要从事深度学习、深度强化学习、离线强化学习算法研究以及在机器人和无人系统中的应用研究
招生信息
欢迎对人工智能、强化学习及机器人方向感兴趣的同学报考！
我们期望你有但不局限于以下专业背景：
计算机专业、自动化专业、人工智能专业、电子信息专业、导航制导专业等。
欢迎对研究方向感谢兴趣的同学联系：lihaoran2015@ia.ac.cn
招生专业
081104-模式识别与智能系统
招生方向
深度强化学习，智能无人系统，具身智能
教育背景
2015-09--2020-08 中国科学院自动化研究所 博士
2011-09--2015-07 中南大学 学士
2011-09--2015-07 中南大学 学士
工作经历
工作简历
2022-05~现在, 中国科学院自动化研究所, 副研究员
2020-10~2022-05,中国科学院自动化研究所, 助理研究员
2020-10~2022-05,中国科学院自动化研究所, 助理研究员
教授课程
计算智能
出版信息
发表论文
[1] Xin Liu, Yaran Chen, Haoran Li, Boyu Li, Dongbin Zhao. Cross-domain Random Pre-training with Prototypes for Reinforcement Learning. IEEE Transactions on Systems, Man, and Cybernetics: Systems[J]. 2025, 第 3 作者null(null): https://arxiv.org/abs/2302.05614.
[2] Xin Liu, Yaran Chen, Haoran Li. Sample-efficient Unsupervised Policy Cloning from Ensemble Self-supervised Labeled Videos. arxiv. 2025, 第 3 作者 通讯作者 null(null): https://arxiv.org/abs/2412.10778.
[3] 陈亚冉, 崔文博, 陈苑文, 李浩然, 谭米宁, 张鑫尧, 刘谨瑞, 赵冬斌, 王鹤. RoboGPT: an LLM-based Long-term Decision-making Embodied Agent for Instruction Following Tasks. IEEE Transactions on Cognitive and Developmental Systems[J]. 2025, 第 4 作者null(null):
[4] 孙敬博, 涂崧峻, 张启超, 李浩然, 刘鑫, 陈亚冉, 陈珂, 赵冬斌. Unsupervised Zero-Shot Reinforcement Learning via Dual-Value Forward-Backward Representation. The Thirteenth International Conference on Learning Representations. 2025, 第 4 作者null(null):
[5] 张启超, 方兴, 徐凯旋, 李浩然, 赵冬斌. High-Quality Synthetic Data is Efficient for Model-Based Offline Reinforcement Learning. International Joint Conference on Neural Networks (IJCNN). 2024, 第 4 作者
[6] 韩煜, 陈亚冉, 李浩然. ATV3D: 3D Object Detection from Attention-Based Three-View Representation. International Joint Conference on Neural Networks. 2024, 第 3 作者
[7] Hu, Guangzheng, Zhu, Yuanheng, Li, Haoran, Zhao, Dongbin. FM3Q: Factorized Multi-Agent MiniMax Q-Learning for Two-Team Zero-Sum Markov Game. IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE. 2024, 第 3 作者http://dx.doi.org/10.1109/TETCI.2024.3383454.
[8] Haoran Li, Zhennan Jiang, Yuhui Chen, Dongbin Zhao. Generalizing Consistency Policy to Visual RL with Prioritized Proximal Experience Regularization. arxiv. 2024, 第 1 作者null(null): https://arxiv.org/abs/2410.00051.
[9] Haoran Li, Shasha Liu, Mingjun Ma, Guangzheng Hu, Yaran Chen, Dongbin Zhao. NeuronsGym: A Hybrid Framework and Benchmark for Robot Tasks with Sim2Real Policy Learning. IEEE Transactions on Emerging Topics in Computational Intelligence[J]. 2024, 第 1 作者null(null): https://arxiv.org/abs/2302.03385.
[10] 李浩然, 张曜程, 温颢玮, 朱圆恒, 赵冬斌. Stabilizing Diffusion Model for Robotic Control with Dynamic Programming and Transition Feasibility. IEEE Transactions on Artificial Intelligence[J]. 2024, 第 1 作者
[11] 李博宇, 李浩然, 朱圆恒, 赵冬斌. MAT: Morphological Adaptive Transformer for Universal Morphology Policy Learning. IEEE Transactions on Cognitive and Developmental Systems[J]. 2024, 第 2 作者 通讯作者
[12] 陈宇辉, 李浩然, 赵冬斌. Boosting Continuous Control with Consistency Policy. International Conference on Autonomous Agents and Multiagent Systems. 2024, 第 2 作者 通讯作者
[13] Hu, Guangzheng, Li, Haoran, Liu, Shasha, Zhu, Yuanheng, Zhao,Dongbin. NeuronsMAE: A Novel Multi-Agent Reinforcement Learning Environment for Cooperative and Competitive Multi-Robot Tasks. 2023 International Joint Conference on Neural Networks(IJCNN). 2023, 第 2 作者 通讯作者
[14] Li, Jiaqi, Li, Haoran, Chen, Yaran, Ding, Zixiang, Li, Nannan, Ma, Mingjun, Duan, Zicheng, Zhao, Dongbing. ABCP: Automatic Block-wise and Channel-wise Network Pruning via Joint Search. IEEE Transactions on Cognitive and Developmental Systems[J]. 2022, 第 2 作者 通讯作者
[15] Li, Haoran, Duan, Zicheng, Li, Jiaqi, Ma, Mingjun, Chen, Yaran, Zhao, Dongbin. Neurons Perception Dataset for RoboMaster AI Challenge. 2022 IEEE World Congress on Computational Intelligence (WCCI). 2022, 第 1 作者
[16] Li, Haoran, Chen, Yaran, Zhang, Qichao, Zhao, Dongbin. BiFNet: Bidirectional Fusion Network for Road Segmentation. IEEE TRANSACTIONS ON CYBERNETICS[J]. 2021, 第 1 作者http://dx.doi.org/10.1109/TCYB.2021.3105488.
[17] Li, Haoran, Zhang, Qichao, Zhao, Dongbin. Deep Reinforcement Learning-Based Automatic Exploration for Navigation in Unknown Environment. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS[J]. 2020, 第 1 作者31(6): 2064-2076, http://dx.doi.org/10.1109/TNNLS.2019.2927869.
[18] Chen, Yaran, Zhao, Dongbin, Li, Haoran, IEEE. Deep Kalman Filter with Optical Flow for Multiple Object Tracking. 2019 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC). 2019, 第 3 作者3036-3041, https://www.webofscience.com/wos/woscc/full-record/WOS:000521353903010.
[19] Zhao Dongbin, Li Haoran, Li Dong, Guo Ping, Chen Yaran. A Temporal-based Deep Learning Method for Multiple Objects Detection in Autonomous Driving. 2018, 第 2 作者http://ir.ia.ac.cn/handle/173211/23521.
[20] Chen, Yaran, Zhao, Dongbin, Li, Haoran, Li, Dong, Guo, Ping, IEEE. A temporal-based deep learning method for multiple objects detection in autonomous driving. 2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN). 2018, 第 3 作者
[21] Lu, Yi, Chen, Yaran, Zhao, Dongbin, Li, Haoran, IEEE. Hybrid Deep Learning Based Moving Object Detection via Motion prediction. 2018 CHINESE AUTOMATION CONGRESS (CAC). 2018, 第 4 作者1442-1447, http://apps.webofknowledge.com/CitedFullRecord.do?product=UA&colName=WOS&SID=5CCFccWmJJRAuMzNPjj&search_mode=CitedFullRecord&isickref=WOS:000459239501089.
[22] Qichao Zhang, Haoran Li, Dongbin Zhao. Comparison of methods to efficient graph SLAM under general optimization framework. YAC 2017. 2017, 第 2 作者*, http://ir.ia.ac.cn/handle/173211/19422.
[2] Xin Liu, Yaran Chen, Haoran Li. Sample-efficient Unsupervised Policy Cloning from Ensemble Self-supervised Labeled Videos. arxiv. 2025, 第 3 作者 通讯作者 null(null): https://arxiv.org/abs/2412.10778.
[3] 陈亚冉, 崔文博, 陈苑文, 李浩然, 谭米宁, 张鑫尧, 刘谨瑞, 赵冬斌, 王鹤. RoboGPT: an LLM-based Long-term Decision-making Embodied Agent for Instruction Following Tasks. IEEE Transactions on Cognitive and Developmental Systems[J]. 2025, 第 4 作者null(null):
[4] 孙敬博, 涂崧峻, 张启超, 李浩然, 刘鑫, 陈亚冉, 陈珂, 赵冬斌. Unsupervised Zero-Shot Reinforcement Learning via Dual-Value Forward-Backward Representation. The Thirteenth International Conference on Learning Representations. 2025, 第 4 作者null(null):
[5] 张启超, 方兴, 徐凯旋, 李浩然, 赵冬斌. High-Quality Synthetic Data is Efficient for Model-Based Offline Reinforcement Learning. International Joint Conference on Neural Networks (IJCNN). 2024, 第 4 作者
[6] 韩煜, 陈亚冉, 李浩然. ATV3D: 3D Object Detection from Attention-Based Three-View Representation. International Joint Conference on Neural Networks. 2024, 第 3 作者
[7] Hu, Guangzheng, Zhu, Yuanheng, Li, Haoran, Zhao, Dongbin. FM3Q: Factorized Multi-Agent MiniMax Q-Learning for Two-Team Zero-Sum Markov Game. IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE. 2024, 第 3 作者http://dx.doi.org/10.1109/TETCI.2024.3383454.
[8] Haoran Li, Zhennan Jiang, Yuhui Chen, Dongbin Zhao. Generalizing Consistency Policy to Visual RL with Prioritized Proximal Experience Regularization. arxiv. 2024, 第 1 作者null(null): https://arxiv.org/abs/2410.00051.
[9] Haoran Li, Shasha Liu, Mingjun Ma, Guangzheng Hu, Yaran Chen, Dongbin Zhao. NeuronsGym: A Hybrid Framework and Benchmark for Robot Tasks with Sim2Real Policy Learning. IEEE Transactions on Emerging Topics in Computational Intelligence[J]. 2024, 第 1 作者null(null): https://arxiv.org/abs/2302.03385.
[10] 李浩然, 张曜程, 温颢玮, 朱圆恒, 赵冬斌. Stabilizing Diffusion Model for Robotic Control with Dynamic Programming and Transition Feasibility. IEEE Transactions on Artificial Intelligence[J]. 2024, 第 1 作者
[11] 李博宇, 李浩然, 朱圆恒, 赵冬斌. MAT: Morphological Adaptive Transformer for Universal Morphology Policy Learning. IEEE Transactions on Cognitive and Developmental Systems[J]. 2024, 第 2 作者 通讯作者
[12] 陈宇辉, 李浩然, 赵冬斌. Boosting Continuous Control with Consistency Policy. International Conference on Autonomous Agents and Multiagent Systems. 2024, 第 2 作者 通讯作者
[13] Hu, Guangzheng, Li, Haoran, Liu, Shasha, Zhu, Yuanheng, Zhao,Dongbin. NeuronsMAE: A Novel Multi-Agent Reinforcement Learning Environment for Cooperative and Competitive Multi-Robot Tasks. 2023 International Joint Conference on Neural Networks(IJCNN). 2023, 第 2 作者 通讯作者
[14] Li, Jiaqi, Li, Haoran, Chen, Yaran, Ding, Zixiang, Li, Nannan, Ma, Mingjun, Duan, Zicheng, Zhao, Dongbing. ABCP: Automatic Block-wise and Channel-wise Network Pruning via Joint Search. IEEE Transactions on Cognitive and Developmental Systems[J]. 2022, 第 2 作者 通讯作者
[15] Li, Haoran, Duan, Zicheng, Li, Jiaqi, Ma, Mingjun, Chen, Yaran, Zhao, Dongbin. Neurons Perception Dataset for RoboMaster AI Challenge. 2022 IEEE World Congress on Computational Intelligence (WCCI). 2022, 第 1 作者
[16] Li, Haoran, Chen, Yaran, Zhang, Qichao, Zhao, Dongbin. BiFNet: Bidirectional Fusion Network for Road Segmentation. IEEE TRANSACTIONS ON CYBERNETICS[J]. 2021, 第 1 作者http://dx.doi.org/10.1109/TCYB.2021.3105488.
[17] Li, Haoran, Zhang, Qichao, Zhao, Dongbin. Deep Reinforcement Learning-Based Automatic Exploration for Navigation in Unknown Environment. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS[J]. 2020, 第 1 作者31(6): 2064-2076, http://dx.doi.org/10.1109/TNNLS.2019.2927869.
[18] Chen, Yaran, Zhao, Dongbin, Li, Haoran, IEEE. Deep Kalman Filter with Optical Flow for Multiple Object Tracking. 2019 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC). 2019, 第 3 作者3036-3041, https://www.webofscience.com/wos/woscc/full-record/WOS:000521353903010.
[19] Zhao Dongbin, Li Haoran, Li Dong, Guo Ping, Chen Yaran. A Temporal-based Deep Learning Method for Multiple Objects Detection in Autonomous Driving. 2018, 第 2 作者http://ir.ia.ac.cn/handle/173211/23521.
[20] Chen, Yaran, Zhao, Dongbin, Li, Haoran, Li, Dong, Guo, Ping, IEEE. A temporal-based deep learning method for multiple objects detection in autonomous driving. 2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN). 2018, 第 3 作者
[21] Lu, Yi, Chen, Yaran, Zhao, Dongbin, Li, Haoran, IEEE. Hybrid Deep Learning Based Moving Object Detection via Motion prediction. 2018 CHINESE AUTOMATION CONGRESS (CAC). 2018, 第 4 作者1442-1447, http://apps.webofknowledge.com/CitedFullRecord.do?product=UA&colName=WOS&SID=5CCFccWmJJRAuMzNPjj&search_mode=CitedFullRecord&isickref=WOS:000459239501089.
[22] Qichao Zhang, Haoran Li, Dongbin Zhao. Comparison of methods to efficient graph SLAM under general optimization framework. YAC 2017. 2017, 第 2 作者*, http://ir.ia.ac.cn/handle/173211/19422.
科研活动
科研项目
（ 1 ） 面向移动机器人的任务规划和技能库算法研究, 负责人, 其他, 2024-05--2024-10
（ 2 ） 多智能体对抗博弈仿真及实体验证, 负责人, 国家任务, 2023-01--2027-12
（ 3 ） 复杂动态环境下基于深度强化学习的机器人导航策略, 负责人, 国家任务, 2022-01--2024-12
（ 4 ） 中国科学院特聘研究助理项目, 负责人, 中国科学院计划, 2022-01--2023-12
（ 2 ） 多智能体对抗博弈仿真及实体验证, 负责人, 国家任务, 2023-01--2027-12
（ 3 ） 复杂动态环境下基于深度强化学习的机器人导航策略, 负责人, 国家任务, 2022-01--2024-12
（ 4 ） 中国科学院特聘研究助理项目, 负责人, 中国科学院计划, 2022-01--2023-12