随着我国城市化脚步的加快和汽车的普及，交通问题己成为不可忽视的全球性问 题。为解决交通问题，许多发达国家都投入大量的人力及物力进行智能交通系统的研究。 智能小车是对智能车辆的一种前期模拟。论文以智能小车为基础，对智能小车的电机以 及舵机控制进行了研究。在电机控制方面，为解决常规PID 控制效果不理想的问题，文 章将模糊控制与PID 控制相结合，构成参数可调整的模糊PID 控制器；在舵机控制方面， 为解决智能小车因经常改变方向导致系统振荡问题，引入微分先行PID 算法，其特点是 只对速度测量值和舵机偏转量进行微分，而不对方向的设定值进行微分。文中分别对电 机及舵机的数学模型进行了Matlab 仿真，仿真结果表明，本文控制算法优于常规PID 控制，不仅响应速度快，而且运行稳定、可靠。为了在DSP28335 上实现上述算法，论 文对图像前期处理以及黑色引导线提取方法也进行了探讨，进而完成了系统硬件设计以 及软件设计，最后对系统进行了调试，试验测试结果表明，采用DSP28335 控制器和 CMOS 摄像头检测方式设计的赛车能够平稳快速的按照指定轨道完成寻迹行驶，具备速 度响应快，稳态误差小、抗干扰能力强等优点，以此为基础，对今后开发从事救灾、危 险环境检测等方面的智能机器人研究具有一定的实用价值。

Traffic problems have become global issues which cannot be ignored with the speeding up of urbanization in our country and the popularity of cars. In order to solve the traffic problems, many developed countries invest a lot of manpower and material resources to research of intelligent transportation system. Smart car is a preliminary simulation of intelligent vehicle. This paper discussed the control method of motor and servo based on smart car. In terms of motor control, in order to solve the problems that the conventional PID control is not ideal, an adaptive fuzzy PID controller is introduced which composed by combining fuzzy control and PID control, and its parameters can be adjusted. In terms of steering gear control, in order to solve the oscillation problems of smart car system caused by changing direction frequently, a differential forward PID controller is introduced. Its characteristic is that only the velocity and the steering gear deflection value get the differential, and the set value of direction does not. The Matlab simulation of algorithm for motor and steering gear mathematical model is implemented. The results show that the algorithm of this paper is superior to the conventional PID, It exerts fast response speed, and running is stable and reliable. In order to realize the algorithm on DSP28335, the method of preprocessing of image and extract of black guide line is discussed. And then completed the system hardware and software has designed. Finally, the system has debugged. The result shows that the designed car, which adopts DSP28335 controller and CMOS camera detection method, can complete tracing according to the orbit, and has the advantage of fast response, small steady-state error and high anti-interference ability. Base on this, we can design rescue robot, dangerous environment inspection robot, and so on, and it has certain practical value.