【作者】 李凡红；

【导师】 董春；

【作者基本信息】 北京交通大学 ， 电气工程， 2010， 硕士

【摘要】 本质不稳定两轮小车是一种特殊轮式移动机器人,其动力学系统具有多变量、非线性、强耦合、参数不确定性等特性,是研究各种控制方法的一个理想平台。其体积小、结构简单、运动灵活,适于在狭小和危险的空间内工作,在民用和军事上有着广泛的应用前景；本课题旨在研制一种两轮自平衡小车。该系统是一种两轮左右平行布置的,像传统的倒立摆一样,本身是一个自然不稳定体,必须施加强有力的控制手段才能使之稳定。其工作原理是系统以姿态传感器(陀螺仪、加速度计)来监测车身所处的俯仰状态和状态变化率,通过高速中央处理器计算出适当数据和指令后,驱动电动机产生前进或后退的加速度来达到车体前后平衡的效果。本文在总结和归纳国内外两轮自平衡小车的研究现状后,在缺少所需的成形的实验对象的情况下,选用适当的控制器、执行电机和传感器,设计出两轮自平衡小车的驱动电路,实现了两轮小车的硬件控制系统。陀螺仪存在漂移的问题及加速度计的动态响应慢,对于系统的姿态检测而言,单独使用陀螺仪或者加速度计,都不能提供有效和可靠的信息来反映车体的实时状态。本文对传感器两者所采集的数据进行了优化处理,补偿陀螺仪的漂移误差和加速度计的动态误差,得到一个更优的倾角近似值。由于实际的小车本体未加工完成,不能在此基础上进行其控制算法的研究。为了提前研究其控制策略同时缩短设计周期,本文选用LEGO机器人为试验平台,搭建所需的小车模型,利用牛顿运动定律对该系统进行数学建模,并对其线性化处理,得到相应的状态方程。在此基础之上详细介绍了两种不同的控制策略,即PID技术和极点配置技术,并通过实验来验证了极点配置方法对该系统有更好的控制效果。更多还原

【Abstract】 The natural unstable two-wheel vehicle is a special kind of wheeled mobile robot. Its dynamics equations are multi-variable, non-linear, serious coupling and uncertain parameters etc. So it is ideal that various control methods are studied on this platform. The robot is small in mechanism and can make a flexible motion. So it is fit for tasks in narrow and dangerous space and has a wide foreground both in civilian and military application. The subject aims to develop a two-wheel self-balancing vehicle. The device of this system is a parallel arrangement of two single wheels, like a traditional inverted pendulum. Because of itself natural instability, it must be exerted strong control to make it stable. It works as follows:system attitude sensors (gyroscopes and accelerometers) monitor body condition and the state in which the pitch change rate, high-speed central processing unit to calculate the appropriate data and instructions, the motor produces forward or backward to achieve the body balancing result.In this paper, studies on self-balancing two-wheel vehicle at home and abroad are summarized. And in the absence of subjects required and formed, we select the appropriate controller, motors and sensors to design two-wheel self-balancing vehicle drive circuit. And it accomplishes the entire hardware control system.Because of gyro drift problems and slow dynamic response of the accelerometer, for gesture detection system, separate usage of gyro or accelerometer can not provide effective or reliable information to reflect the real state of its body. In this paper, the data collected by the two sensors are optimized to compensate the drift error of gyroscope and accelerometer dynamic error so that the inclination of its body can be approximated better.Since that the real vehicle body is not accomplished, the control algorithms can not be studied on this basis. In order to study the control strategy in advance and shorten the design cycle, the paper has to take advantage of the LEGO robot as test platform, build vehicle model required, get mathematical modeling of the system by Newton’s laws and obtain the corresponding state equation through its linearization. On this basis, the two different control strategies, namely, PID technology and pole placement technique are described in details, and finally the pole configuration is verified that it has better control effect through experiments.更多还原