头像
更新时间:2025.04.29
总访问量:10

梁定坤

| 博士 副研究员 请选择

单位:

职务:

研究方向:

办公地址: B310

办公电话:

电子邮箱: liangdk@zjut.edu.cn

手机访问
  • 个人简介

      梁定坤,南开大学工学博士,浙江工业大学信息工程学院,朝晖特聘副研究员,曾为之江实验室高级研究专员、双足机器人控制项目负责人、机器人中心控制组副组长

      主持之江实验室重大项目经费体量千万、浙江省自然科学青年基金等,参与重点研发、国自然基金等项目,在IEEE TIITIE一区顶刊和国际会议上发表SCI/EI论文20余篇(1/2区顶刊11篇,ESI高被引1篇)受理/授权发明专利10余件,获得全国挑战杯国赛一等奖、天津市优秀个人、小平科技创新团队核心成员)、研究生国家奖学金南开十杰、之江实验室机器人中心先进工作者等荣誉。



      ★研究方向:人形机器人、基于强化学习的运动生成、仿生机械臂、非线性控制 


      ★教育及工作经历

      ★个人简介


  • 科研项目

    ★主要项目经历:

    1结题22.4-24.2之江实验室重大项目,高性能双足机器人及实景化应用-机器人本体运动控制系统,1026万,主持

    项目简介:建立双足机器人高效准确的刚柔耦合多体系统动力学模型,构建大规模、全方位、多工况的动作规划系统,研发速度快、稳定性高、环境适应性强的实时控制算法

    2(在研23.1-25.12浙江省自然科学青年基金,安全交互场景下柔性仿人操作机器人高效跟踪与智能柔顺方法研究,主持

    项目简介:研究仿人操作臂的刚柔耦合动力学建模与最优轨迹规划、复杂干扰下的高效跟踪与实时扰动抑制、力、位、刚度一体的智能柔顺控制

    3在研23.9-26.2工信部高质量发展专项机器人专用控制**1000.2964万,主要参与

    项目简介研究机器人轨迹规划、力位混合控制等算法

    4结题22.2-24.1之江实验室重大项目,输尿管软镜操作机器人系统,1800万,骨干参与

    5结题22.1-24.1之江实验室开放课题,基于强化学习的双足机器人多运动模式与抗扰控制方法研究,项目对接人

    6结题22.1-24.1之江实验室开放课题,柔性仿人手臂轨迹跟踪与扰动抑制技术研究,技术联系人

    7结题19.6-22.11国家重点研发计划重点专项课题,《面向动态生产过程的机器人建模与精准控制》,骨干参与

    8结题20.5-22.5辽宁省自然基金资质计划,《仿生软体机器人建模与智能控制方法》,骨干参与



    ★主要研究工作:

    1、人形机器人平台搭建与应用演示

        研发人形机器人的高速动态行走算法、平衡控制与扰动抑制算法、拟人技能学习方法、全身协调柔顺方法、全身动力学奔跑控制方法等一系列规划与控制方法,完成算法框架梳理、模型搭建、仿真平台开发、嵌入式代码部署以及算法落地应用调试,前往余姚机器人峰会、机器人年会、世界机器人大会上参展,支持亚运会开场、亚残运会宣传片录制。


    (之前合作研发的样机与浙工大最新样机)


    2、人形机器人高速动态行走

        基于角动量提出一种高速动态行走算法,样机最高运动速度突破10km/h



    3、人形机器人平衡抗扰与地形自适应

        基于虚拟力、MPCWBC方法,可抵抗随机干扰和碰撞,适应6种地面(地毯、水泥地、草地、木板、展厅瓷砖等),实机常规速度2km/h(仿真>6km/h),实现不平整路面行走(随机地面高度起伏2cm以上),踩跷跷板(高度10cm以上)维持自身平衡。


    (左:不同地形行走,右:跷跷板平衡)

    (全身力控算法,抵抗外力干扰,实现不平整路面行走)


    4、人形机器人全身协调柔顺

        基于全身协调柔顺方法,减小足地冲击,抵抗外力干扰(负载58kg,3kg球撞击),实现全向柔顺行走,支持连续上下5cm台阶(实机关节范围限制)。



    5、人形机器人基于强化学习的运动生成

        基于深度强化学习策略,设计考虑变速奔跑任务和稳定、能耗等指标的奖励函数,实现不同步频调节,仿真中最高平均运速度突破10km/h,该部分工作已发表至中科院二区期刊Multibody System Dynamics。基于对抗运动先验的强化学习策略,仿真中可实现走、跑、跳、后空翻等复杂拟人动作,同时具备随机干扰下的鲁棒性;通过域随机化等手段探索sim2real。


    (基于DRL的变速奔跑运动生成与Sim2Real)

    (基于对抗运动先验的拟人化运动学习)


    6、人形机器人拟人动作模仿与操作技能学习

        开展人形机器人双臂拟人化运动模仿与操作技能学习算法研发与实验调试,实现任意人类动作演示的模仿,保证动作平滑性,相关工作已发表至中科院二区TOP期刊IEEE Robotics and Automation Letters。


















    人形机器人全身协调与平衡控


  • 科研成果

    主要学术成果:



    1.D Liang(梁定坤), et al. Fuzzy-sliding mode control for humanoid arm robots actuated by pneumatic artificial muscles with unidirectional inputs, saturations, and dead zones, IEEE Transactions on Industrial Informatics, 2022, 18(5): 3011-3021. [中科院一区TOP,影响因子11.7]

    2.D Liang(梁定坤), et al. Energy-based motion control for pneumatic artificial muscle-actuated robots with experiments, IEEE Transactions on Industrial Electronics, 2022, 69(7): 7295-7306. [中科院一区TOP,影响因子7.5]
    3.N Sun(导师), D Liang(梁定坤), et al. Adaptive control for pneumatic artificial muscle systems with parametric uncertainties and unidirectional in put constraints, IEEE Transactions on Industrial Informatics, 2020, 16(2): 969-979. [ESI高被引,中科院一区TOP,影响因子11.7]
    4.Y Wu, P Zhao, D Liang*(通讯作者), et al. A learning-based anti-swing trajectory refinement approach for UAVs with cable-suspended payload without offline training,IEEE Transactions on Intelligent Vehicles, 2024, DOI: 10.1109/TIV.2024.3391788. [中科院一区,影响因子14]
    5.D Liang(梁定坤), et al. Trajectory planning-based control of underactuated wheeled inverted pendulum robots, Science China Information Sciences, 2019, 62(5): 50207. [中科院二区,影响因子7.3]
    6.L Tang, D Liang*(通讯作者), et al. Modeling and reinforcement learning-based locomotion control for a humanoid robot with kinematic loop closures, Multibody System Dynamics, 2024, in press, DOI: 10.1007/s11044-024-10035-z. [中科院二区,影响因子2.6]
    7.B Zheng, D Liang(梁定坤), et al. Frame-by-frame motion retargeting with self-collision avoidance from diverse human demonstrations, IEEE Robotics and Automation Letters, 2024, 9(10): 8706-8713. [中科院二区TOP,影响因子4.6]
    8.Q Qiu, W Wang, H Ying, D Liang(梁定坤), et al.SelFLoc: Selective feature fusion for large-scale point cloud-based place recognition, Knowledge-Based Systems, 2024: 111794. [中科院一TOP,影响因子7.2]
    9.Y Zhou, Y Wu, D Liang(梁定坤),et al. A neural network-based state-constrained control strategy for underactuated aerial transportation systems within narrow workspace, Symmetry, 2024, 16(11): 1512. [JCR Q2,影响因子2.3]
    10.H Gao, Q Qiu, H Liu, D Liang*(通讯作者),et al. ERPoT: Effective and reliable pose tracking for mobile robots based on lightweight and compact polygon maps,IEEE Transactions on Robotics, 2024, 在审. [中科院一区TOP,影响因子9.4]
    11.Y Wu, T Yu, Q Zhang, D Liang(梁定坤),et al. Task-oriented finite-time adaptive sliding mode-based antiswing control for aerial cargo transportation systems with underactuated swing motion suppression,Control and Decision, 2024, 在审. [国内控制领域著名期刊,EI]
    12.Y Chen, N Sun(导师), D Liang(梁定坤), et al. A neuroadaptive control method for pneumatic artificial muscle systems with hardware experiments, Mechanical Systems and Signal Processing, 2021, 146: 106976-1-15. [中科院一区TOP,影响因子7.9]
    13.G Liu, N Sun(导师), D Liang(梁定坤), et al. Neural network-based adaptive command filtering control for pneumatic artificial muscle robots with input uncertainties, Control Engineering Practice, 2022, 118: 104960. [中科院二区,影响因子5.4]
    14.Y Wu, N Sun(导师), Y Fang(导师), D Liang(梁定坤). An increased nonlinear coupling motion controller for underactuated multi-TORA systems: Theoretical design and hardware experimentation, IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019, 49(6): 1186–1193. [中科院一区TOP,影响因子8.6]
    15.梁定坤, 等人. 采用扰动估计的气动人工肌肉系统非线性控制, 控制理论与应用, 2019, 36(11): 1912-1919. [国内控制领域顶刊,EI]
    16.梁定坤, 等人. 气动人工肌肉驱动的机器人控制方法研究现状概述. 控制与决策, 2021, 36(1): 27–41. [国内控制领域著名期刊,EI]
    17.D Liang(梁定坤), et al. Dynamic modeling and analysis for dual pneumatic artificial muscle actuated manipulators. Proceedings of the 2019 IEEE/ASME International Conference on Advanced Intelligent Mechatronics, Hong Kong, China, July 8-12, 2019: 691-696. [EI 检索]
    18.G Gao, …, D Liang*(通讯作者), et al. Hierarchical optimal motion planning for piano playing robots with dexterous fingers. Proceedings of the 2022 IEEE International Conference on Robotics and Biomimetics, Xishuangbanna, China, Dec. 5-9, 2022: 357-362. [EI 检索]
    19.D Liang(梁定坤), et al. Dynamic modeling and control of inverted pendulum robots moving on undulating pavements. Proceedings of the Seventh International Conference on Information Science and Technology, Da Nang, Vietnam, April 16-19, 2017: 115-120. [EI 检索]
    20.D Liang(梁定坤), et al. Modeling and motion control of self-balance robots on the slope. Proceedings of the 31st Youth Academic Annual Conference of Chinese Association of Automation, Wuhan, China, Nov. 11-13, 2016: 93–98. [EI 检索]
    21.D Liang(梁定坤), et al. Differential flatness-based robust control of self-balanced robots. IFAC-PapersOnLine, 2018, 51(31): 949-954. [EI]
    22.Y Huang, X Wang, D Liang(梁定坤), et al. Modeling and parameter identification for human-robot coupled systems in powered lower limb prostheses. Proceedings of the 2023 World Robot Conference, Beijing, China, Aug. 19-21, 2023: 14-20. [最佳论文提名, EI 检索]
    23.Y Liu, …, D Liang(梁定坤), et al. Design and development of the dual arm robot DARO. Proceedings of the 2023 World Robot Conference, Beijing, China, Aug. 19-21, 2023: 97-102. [EI 检索]
    24.D Liang(梁定坤), et al. Time optimal planning-based high-speed running motion generation for humanoid robots with safety constraints. Proceedings of the 2023 Chinese Control Conference, Tianjin, China, July 24-26, 2023: 3017-3023. [国内控制顶级会议,EI 检索]
    25.L Tang, D Liang*(通讯作者), et al. A rigid-flexible coupling recursive formulation for dynamic modeling of biped robots. Proceedings of the 16th International Conference on Intelligent Robotics and Applications, Hangzhou, China, July 5-7, 2023: 69-80. [EI 检索]
    26.Y Zhu, …, D Liang*(通讯作者), et al. Fast walking of position-controlled biped robot based on whole-body compliance control. Proceedings of the 16th International Conference on Intelligent Robotics and Applications, Hangzhou, China, July 5-7, 2023: 344-354. [EI 检索]
    27.C Gao, …, D Liang*(通讯作者), et al. Whole body balance control for bipedal robots based on virtual model control. Proceedings of the 16th International Conference on Intelligent Robotics and Applications, Hangzhou, China, July 5-7, 2023: 355-367. [EI 检索]
    28.J Xiong, D Liang(梁定坤), et al. Nonsmooth dynamic modeling of a humanoid robot with parallel mechanisms. Proceedings of the 16th International Conference on Intelligent Robotics and Applications, Hangzhou, China, July 5-7, 2023: 294-308. [EI 检索]
    29.J Tang, …, D Liang(梁定坤), et al. Omnidirectional walking realization of a biped robot. Proceedings of the 16th International Conference on Intelligent Robotics and Applications, Hangzhou, China, July 5-7, 2023: 258-269. [EI 检索]
    30.朱怡婷, 梁定坤, 等人. 一种应用于双足机器人快速行走的全身柔顺控制方法, 中国发明专利, 专利号: 202310009303.9, 授权.
    31.谢安桓, 梁定坤, 等人. 步态模仿方法、装置、电子装置和存储介质, 中国发明专利, 专利号: 202211738815.0, 实审.
    32.孙宁, 梁定坤, 方勇纯, 等人. 基于关节角度和肌肉长度测量的机器人控制系统及方法, 中国发明专利, 专利号: 202110384674.6, 授权.
    33., 王鑫, 梁定坤, 等人. 一种仿人机器人的行走控制方法、装置及存储介质,中国发明专利, 专利号:202311337925.0, 受理.
    34.桑文闯, 梁定坤, 等人.一种基于强化学习的双足机器人步态控制方法及系统,中国发明专利,专利号: 202311736353.3, 受理.
    35.朱怡婷, 梁定坤, 等人. 一种应用于双足机器人快速行走的全身柔顺控制方法, PCT国际专利, 专利号: PCT/CN2023/125001, 受理.


链接

更新时间:2025.04.29
总访问量:10