学术论文

      基于生物反射模型的四足机器人坡面运动控制与越障研究

      Study on quadruped robot ramp-trotting and obstacle-crossing based on biological reflex model

      摘要:
      为了提高四足机器人在包含坡面和障碍物等复杂地形中的运动能力与环境自适应能力,在对四足机器人基本步态研究的基础上,利用生物节律运动和反射控制机理,对四足机器人的适应性行走控制模型进行了研究.建立了适用于四足机器人坡面运动以及越障运动的前庭反射和屈肌反射数学模型,根据该数学模型构建的生物反射控制器与机器人膝、髋关节CPG控制网络有机融合,构成了协调性好、整体性高的控制系统.通过Adams/Matlab联合仿真,验证了所提出控制模型的可行性与有效性.该模型能够有效地使前膝后肘式四足机器人流畅、平稳地完成上下坡运动,并具备自适应越障运动能力.
      Abstract:
      To improve the locomotion ability and terrain adaptive ability of a quadruped robot through complex terrains including ramps and obstacles, based on the study of quadruped robot basic gaits, a biomimetic adaptive control model for the quadruped robot is studied with the mechanisms regarding animals' rhythmic locomotion and reflex control.The mathematical models of vestibular and flexor reflexes are established to realize trotting on a ramp and obstacle crossing.Reflex controllers derived from the aforementioned mathematical models are integrated with the central pattern generator (CPG) networks of knee and hip joints of the robot, so that a control system is created with good coordination and high degree of integrity.By co-simulation of Adams and Matlab, the feasibility and effectiveness of both the vestibular reflex model and the flexor reflex model are verified.Under the control of such control model, the quadruped robot with all outward knee joints can stably trot uphill or downhill and adaptively achieve obstacle avoidance and crossing.
      作者: 葛卓 [1] 罗庆生 [1] 贾燕 [2] 李华师 [2]
      Author: Ge Zhuo [1] Luo Qingsheng [1] Jia Yan [2] Li Huashi [2]
      作者单位: 北京理工大学机电学院,北京,100081 北京理工大学机械与车辆学院,北京,100081
      年,卷(期): 2017, 47(4)
      分类号: TP242.6
      在线出版日期: 2017年8月15日
      基金项目: 国家部委预研基金资助项目