电磁驱动无人驾驶机器人多场耦合机理及仿生集成优化研究

The project's purpose is to solve the problems of multi-physics coupling effects among structure-displacement field, electromagnetic field, temperature field, and global performance optimization for linear electromagnetic-drived robot driver. The research contents are as follows. Firstly, the electromagnetic linear actuators mathematical model and the coordinated control model of multiple manipulators with intelligent decision-making ability are established. During the process of modelling, the influence relationship among electronic subsystem, magnetic subsystem, thermal subsystem and mechanical subsystem is considered. Both of the models are satisfied to achieve parameter adjustment fast and accurately, in order to meet the requirements of high control response and high control precision for linear electromagnetic drive. On the basis of this, the multi-field coupling dynamic model combined with structural field, electronic field, magnetic field and thermal field is established. The quantitative relationship between the system structure factors including structure parameters and manufacture precision, and its electronic performance including control response and control precision is determined by the model. And the inner relationship among the fields is discovered. Finally, based on multidiscipline parallel design optimization, the project explores structure-control integrated optimization for motion performance optimization and design parameters matching. This is used by modern intelligent bionic algorithm, such as particle swarm optimization, artificial bee colony, differential evolution algorithm, and so on. The interaction between structure parameter and control parameter is analyzed. The project will help to further reveal the effecting mechanism between the system components and overall performance, and the interaction mechanism between structure parameter and control parameter for electromagnetic-drived robot driver. It can provide a theoretical basis for its application to unmanned aerial vehicles, unmanned weapons platforms, battlefield reconnaissance vehicles, space robots, planetary roaming vehicles, and other fields.

针对电磁驱动无人驾驶机器人的结构位移场、电磁场、温度场等多物理场间存在耦合效应和系统全局性能最优问题，综合考虑电、磁、热和机械等各个子系统间的相互影响关系，建立电磁直线执行器数学模型及具备智能决策能力的多机械手协调控制模型，以满足应用电磁驱动后的高精度和高速度要求，实现快速、准确的参数调节；在此基础上，建立机、电、磁、热等多物理场耦合的动力学模型，确定系统结构因素与其响应及精度等性能间的定量关系，发掘各场之间的内在联系；最后基于多学科并行设计优化的思想，探索应用鸟群、蜜蜂群、微分进化算法等现代智能仿生算法对结构和控制同时进行运动性能优化和设计参数匹配，分析两者之间的相互影响。通过本项目的研究，可进一步揭示电磁驱动无人驾驶机器人系统各组成部分对整体性能的作用机制及结构与控制间的相互影响机理，为其在无人驾驶车辆、无人驾驶武器平台、战地侦察车、太空机器人、星球探测漫游车等领域的应用提供理论基础。

无人驾驶机器人是一种无需对车辆进行改装，可无损安装在驾驶室内，替代驾驶员在恶劣条件和危险环境下进行自动驾驶的智能化机器人。由于具有良好的通用性、无损性、可重复性、安全性、可靠性以及准确性，因此它可广泛适用于民用和军用领域。无人驾驶机器人是进行汽车试验以及实现无人驾驶的一种比较理想的解决方案。电磁驱动无人驾驶机器人采用电磁直线执行器将电能直接转换为需要的直线形式的机械能，以控制油门机械腿、离合器机械腿、制动器机械腿、换挡机械手等执行机构动作，消除了机械机构的迟滞，提高了整体系统动态响应，并能使系统结构更加紧凑。本项目首先根据无人驾驶机器人的驱动特性，综合考虑电场方程、磁场方程、电磁力方程和机械结构动力微分方程，建立了电磁直线执行器数学模型，实现了电磁直线执行器输出速度和驱动力的跟踪控制，并融合驾驶员操纵驾驶规律，提出了一种基于模糊逻辑理论的电磁驱动无人驾驶机器人递阶控制模型体系结构，实现了无人驾驶机器人多机械手的协调控制。在此研究基础上，构建了电磁驱动无人驾驶机器人机、电、磁、热等多物理场耦合的动力学模型，分析了电磁直线执行器极弧系数、电机级距等结构参数与无人驾驶机器人运动响应和精度的内在联系，研究表明合适的电磁直线执行器结构参数有利于减少电磁力的波动，从而有利于提高无人驾驶机器人的运动精度。最后，基于模拟退火粒子群仿生优化算法，进行电磁驱动无人驾驶机器人机械结构参数和控制参数的集成优化，研究表明仿生集成优化的控制参数能使控制器能够精确快速地跟踪无人驾驶机器人期望运动轨迹，跟踪误差小于0.35mm，仿生集成优化的电磁驱动无人驾驶机器人结构参数能够实现换挡机械手的选挡轨迹误差和挂挡轨迹误差均小于2mm，并能够使无人驾驶机器人机械腿的驱动电磁直线执行器的行程有显著降低。该项目的研究成果可为电磁驱动无人驾驶机器人在环保汽车制造、无人驾驶汽车、无人武器平台、战地侦察车、太空机器人、星球探测漫游车等领域的应用提供理论基础。