3-D双足机器人研究所涉及的动力学与控制学科领域难点很多，如高维非线性、随机不确定、欠驱动、多域混杂、非光滑，也包含时变或时滞等问题。本项目拟在对平面双足机器人研究基础上，进一步开展对3-D双足机器人步态规划与多目标优化以及随机不确定性鲁棒控制研究。发展3-D双足机器人步态切换过渡阶段与步内稳定性控制策略，研究机器人切换步态的稳定性控制方法；深入分析双足机器人动态行走的多重性能要求，结合混杂零动力学方法，发展欠驱动3-D双足机器人的多目标优化算法；建立欠驱动3-D双足机器人的随机非线性动力学模型，对机器人控制器进行设计和优化，发展机器人动态行走的鲁棒控制策略；搭建先进的3-D双足机器人实验平台，开展3-D双足机器人动态行走的实验研究，验证上述理论与方法的可行性以及仿真计算结果的正确性。本项目的研究对于今后进一步设计与研制鲁棒性更强、效率更高的3-D双足机器人具有重要的科学意义和参考价值。

In a biped robot, there exist many difficulties in the field of dynamics and control, such as high dimension and nonlinearity, random uncertainty, underactuation, multi-domain hybrid, non-smoothness, and even time-variant or time-delay problems can arise. Based on our previous work on planar biped robot, this project will continue to study the gait planning, multi-objective optimization, and robust control on random uncertainty of a 3-D biped robot. We will propose the stability control strategy on the gait transition and in-step stages, and study the stability control method of transitional gait of the 3-D biped robot. By analyzing multiple performance demands of the robot in bipedal walking and combing with the method of hybrid zero dynamics, the multi-objective optimization algorithm will be developed. A stochastically nonlinear dynamical model will be established to design and optimize the controller in the underactuated 3-D biped robot and develop the robust control strategy on its stable walking. A physical prototype platform of the biped robot will be set up to perform the bipedal walking experiments to validate the theoretical and simulational results. We believe that there are important scientific significance and practical values to develop and manufacture a 3-D biped robot with stronger robustness and higher efficiency by this project.