基于AVR单片机软件控制系统设计的机器人毕业论文外文翻译.docx

《基于AVR单片机软件控制系统设计的机器人毕业论文外文翻译.docx》由会员分享，可在线阅读，更多相关《基于AVR单片机软件控制系统设计的机器人毕业论文外文翻译.docx（19页珍藏版）》请在三一办公上搜索。

1、外文资料The Software Control System Design for Robot Based on AVR MCUKai Yang， Junmei Zhang， Wenbin LI， Liu YangSchool of TechnologyBeijing Forestry UniversityBeijing， ChinaCorresponding author： Junmei Zhang， AbstractWith the fast development of new principle in intelligent control， mobile robot has bec

2、ome the focus in the field of robotics and automation for scholars. Robot competitions are active worldwide and expanded. Based on the competition requirements and rules of 2008 Asia-Pacific Robot Contest， the control software of the competition robot is designed in this paper. In the control system

3、， the AVR single chip computer and C-language are used to achieve the DC motor control， rudder servo control， and path planning of the robot. From an overall perspective， the structure of the program is clear and could meet the control requirements of the competition robot. The program can also meet

4、 the expected function.Keywords- robot； control software； AVR SCM； path planningI. INTRODUCTION Robot is the highest manifestation of artificial intelligence and product of the development of automation technology. Robot competitions are active worldwide and expanded. This article is based on the 20

5、08 Asia-Pacific Robot Contest launched on robotics research. The task of the robot competition requirements and rules of the game is that the robot track on the path line to snatch at the object which is put on the top on the pillars. This designs control request in the standard site is that the rob

6、ot can track the signal line steady and fast， arrives at specified location according to the predetermined way and realizes the snatch and so on a series of movements assigned 1. Robot control system itself is complex. This article mainly involves software control. The whole control system is divide

7、d into the following tasks： DC motor control， multichannel rudder control， path planning and so on.II. DC MOTOR CONTROLA. The choice of control schemeThis design uses two DC motor as a system driver to control the movement of the robot left and right wheels. MCU control methods for motor are general

8、ly two types： analog control and digital control methods. Digital control method is to issue specific instructions to the motor control unit directly through the serial port of MCU in order to achieve the corresponding control requirements. The motors currently on the market are equipped with approp

9、riate professional control module. The control module is one that contains the instruction decoding， PID adjustment， photoelectric encoder and other modules， with good anti-interference and control advantages of high accuracy2. Because of conditions， such as restrictions on the system board， conside

10、red from the stability and reliability of the overall system， this design chooses the digital control method.B. MCU serial communication theory and flowThe motor control commands are stored in a one dimensional array. Each time one controls the motor， which is to send an array of characters to the c

11、orresponding controller.This work is completed by the AVR controlling universal synchronous and asynchronous serial receiver and transponder (USART) of chip. In this process， the control chip and motor control units use the way of asynchronous serial communication to carry on the correspondence. Con

12、trol chip is as the host computer， while the motor control unit is as a data receiving terminal. The format of Character frame is 8 data bits， 1 stop bit， no parity bit， and the baud rate is the 9600 bit/s. As the USART interior includes a baud rate generator， its clock input is 8 frequency division

13、 or 16 frequency division of system clock input(depending on the asynchronous communication working pattern， when for double speed pattern is 16 frequency division， otherwise is 8 frequency division). The data transmission selects query mode to carry on. After sending data due out to buffer register

14、， buffer register empty flag bit is kept sending the query. When this flag bit is 1， it means the transmission has completed and next data can be transmitted. The Fig. 1 is the flowchart of serial communication.III. MULTICHANNEL RUDDER CONTROLAs the robot arms multiple joints need to control by the

15、rudder， the MCU must be able to control multichannel rudder. From the correlation between pulse width and angle of rudder， we can see that the pulse width is very short relative to the pulse period and the time of pulse emergence is not necessarily. As long as in a pulse cycle， pulses appearing at a

16、ny time are able to meet the requirement. Therefore， we can control a multichannel rudder by means of generating different output pulses at different times in a signal cycle 3. Below taking timer overflow interrupt which generates PWM to control 6 rudders (this system only uses 2 rudders) simultaneo

17、usly. We suppose the control signal output ports of 6 rudders are respectively portc.0 to portc.5. The time of high level is respectively Th0 to Th5. The control step is as follows： Pull portc.0 first， timer timing Th0. When the timer overflows and causes interrupt， portc.0 is pulled down， while por

18、tc.1 is pulled and timer times Th1.Do this process again and again until portc.5 is pulled down. Then the timer times 20ms-(Th0+ Th1+ Th2+ Th3+ Th4+ Th5) and completes a signal cycle timing. The flowchart of rudder control value calculating program and the rudder control program see the Fig. 2 and F

19、ig. 3 below.IV. PATH PLANNINGWhether robot could walk in accordance with pre-set routes， the key is that the robot could take a straight line and change direction accurately according to control requirement， and continue to walk along straight line after turning the direction. Without appropriate al

20、gorithms， the robot will walk a noticeable s-shaped curve， not only affecting the speed， but also very easy to deviate from the leading line. Robot movement was divided into the following sequence of actions： seeking the signal line to go straight， calculating the number of grid to position， and cha

21、nging direction.A. Line trackingThis design uses three sensors S1， S2， and S3 to ensure the robot to go straight. These three sensors distribute in the front of the robot by the triangular pattern. When a robot moves forward， its status of going forward is determined through the sensor feedback valu

22、es of these three sensors. When the robot has the deflection， speed commands of motor are carried out to produce speed difference of motor of both sides and cause the robot to turn back to guide online in the opposite direction. As the robot speed is quick， it is very easy to appear that three senso

23、rs deviate together from the leading line， which is unable to adjust. Therefore we design state storage variable s_flag to record adjustment process. When the robot deviates from the leading line to the left， MCU makes an order of speed difference to motor control module to make the robot back to th

24、e right side， and assigns a value to the state variable at the same time (e.g. 1). Similarly， when the robot deviates in the opposite direction， it will implement an appropriate action， except for assigning another value (e.g. 0) to the state variable to show differences 3. In the adjustment process

25、， the robot has deviated in certain angle because of the adjustment lag， although having made an order to the motor. As mentioned above， all three sensors deviate from the leading line. Then， robots will make a greater adjust with the speed difference based on the state variable. The adjustment proc

26、ess is that robots will be made a similar action except for increasing the speed difference of adjustment (stop a motor completely) to make robot back to the leading line， when the state variable is 1， which indicates that the robot has make an adjustment about the deviation to its left， but the spe

27、ed difference is not enough to make the robot back to the track. When the state variable is 0， adjustment process can be analogized. The detail line tracking flowchart is shown in Fig. 4.B. Line CountingRobots positioning is accomplished through counting line， namely completing through recording the

28、 cross intersection number the robot passes. This process is also achieved through the sensor. Counting line sensors are installed in the robots distant side of the axis to prevent the straight leading line from interfering counting line sensors due to frequent adjustments robot deflection causes. I

29、n this process， the algorithm choice is also very important. As the signal line width reaches 30mm and scanning is fast， it is possible to repeat counting if the timer counts every time the sensor sweeps the black line. Sensors value will not jump on 30mm signal line. If we adopt the rising or falli

30、ng edge counting， the timer counts only when the sensor value jump from 0 to 1 or from 1 to 0， so we can prevent the emergence of iteration count. The flowchart of the line counting is shown in Fig. 5.C. Turning controlRobots turning is specified by setting two motors at the same rate， so that the r

31、obot revolves around the axe midpoint of the two motors. Robots direction and speed of turning are determined by motors. The control of turning angle is realized by the front-end counting line sensor. Take 90-degree turning for example. When the robot begins to turn， it keeps scanning the front-end

32、sensor at the same. When the sensor scans signal line， which indicates the turning is in the right place， motors stop turning. As whether the turning is in place is determined by whether the front-end sensor has detected the signal line， the time to begin to scan is also very important. If the senso

33、r is above the signal line when starting scanning， the robot would not turn. Therefore， it should be delayed for some time before scanning. It is better to delay until the motor turns half of the angle. The concrete time needs to test according to the speed. When the turning is completed， the robots

34、 posture would have some deflection due to inertia. When the robot walks straight once again， the adjustment needed is relatively large， so it is necessary to make minor adjustment. The adjustment is actually line-tracking process under low speed. The flowchart of the turning control is shown in Fig

35、. 6 and the main flowchart is shown in Fig. 7.V. CONCLUSIONBased on the competition requirements and rules of 2008 Asia-Pacific Robot Contest， the control software of the competition robot is designed in this paper. In the control system， the AVR single chip computer and C-language are used to achie

36、ve the DC motor control， rudder servo control， and path planning of the robot. From an overall perspective， the structure of the program is clear and could meet the control requirements of the competition robot. The control system is precise and steady， which testifies the effectiveness of this meth

37、od in the paper.1 Requirements and rules of 2008 Asia-Pacific Robot Contest. Available： http：/2 Yang Shang-Kuo， “Design and implementation of digital AVR for medium and small synchronous generators，” International Journal of Electrical Engineering， v 12， n 1， pp. 1-13， February 2005.3 Yasir Muhammad

38、， “Adaptive differential pulse code modulation using an AVR microcontroller，” Electronics World， v 115， n 1880， pp. 17-19， August 2021.4 Fryza， T.， “Basic C code implementations for AVR microcontrollers，” 2007 IWSSIP and EC-SIPMCS - Proc. 2007 14th Int. Workshop on Systems， pp.434-437， 2007.5 Butt，

39、Atif Waqar， Iqbal， Javaid1， Bano， “Fabrication and control of 4-dof， autonomous robotic ARM using low cost AVR controller，” Proceedings of the Eight IASTED International Conference on Control and Applications， v 2006， pp. 118-122， 2006.6 Livingston， David， “A controller for robotics and microcontrol

40、ler applications instruction，” ASEE Annual Conference and Exposition， Conference Proceedings， 2007.中文译文基于AVR单片机软件控制系统设计的机器人摘要随着智能控制快速发展的新原则，移动机器人在该范畴已成为关注的焦点学者的机器人和自动化。机器人比赛活跃在世界各地和扩大。基于竞争的2008年亚太机器人大赛的要求和轨则，在这个设计的竞赛机器人控制软件纸张。在控制系统中，对AVR单芯片计算机和C语言是用来实现直流电机的控制，舵伺服控制和路径规划的机器人。从整体的角度来看，结构的轨范是明确的，能够满足竞赛

41、机器人的控制要求。该计划也可以达到预期的功能。关键词机器人；控制软件； AVR单片机；路径规划一.简介机器人是人工智能的最高体现自动化技术的发展和产品。机器人竞赛是活跃在世界各地，并扩大。这本文基于2008年亚太机器人大赛推出机器人研究。机器人的任务比赛要求和游戏轨则是，抢夺的对象，这是把机器人的道路上线轨道顶部的支柱。这种设计的控制要求标准的网站，该机器人可以跟踪信号线稳定快，到达指定的位置，按照预定的方式，实现了一系列的抢夺等分派的改观。机器人控制系统本身是复杂。本文主要涉及软件控制。该整个控制系统分为以下任务：直流多舵控制，电机控制，路径规划和等。二直流电动机的控制A. 控制方案的选择这

42、个设计使用两个DC电机作为系统驱动器控制机器人的运动的左和右轮。单片机控制电机的方式一般有两种：模拟控制和数字控制方式。数字控制方式是发出直接通过电机控制单元的具体指示单片机的串行端口，以实现相应的控制要求。目前市场上的电机配备相应的专业控制模块。该控制模块是一个包含指令译码，PID调节，光电编码器等模块，与良好的抗干扰性和控制精度高的优点。由于在系统上的限制条件下，如板，从稳定性和可靠性的考虑整个系统中，这样的设计选择的数字控制方式。B.单片机的串口通信的理论和流量电机控制的命令被存储在一个一维的数组。每次一控制电机，这是发送一个字符数组中的相应的控制器。完成这项工作是由AVR控制通用同步和

43、异步串行接收器和转发器(USART)芯片。在此过程中，控制芯片和电机控制单元采用异步串行方式通信进行通信。控制芯片作为主计算机，而将电机控制单元作为数据接收终端。字符帧的格式是8位数据位，1位停止位，无奇偶校验位，波特率为9600比特/秒。如USART内部包孕一个波特率发生器，它的时钟输入是8分频或16分频系统时钟输入(取决于异步通信工作模式，以双倍的速度时模式是16分频，否则是8个频师)。数据传输模式进行选择查询。在发送数据缓冲寄存器，缓冲寄存器空符号位不停地发送查询。当此符号位为1，这意味着传输已经完成，下一个数据可以是传输。图1是串行的流程图沟通。三多通道标的目的舵控制作为由需要控制的机

44、器人手臂的多关节舵，MCU必需能够控制多声道舵。从脉冲宽度和舵角之间的相关性，我们可以看到，很短的脉冲宽度相对于脉冲周期和脉冲泛起的时间是不纷歧定。只要在一个脉冲周期，脉冲泛起在任何时间都能够满足要求。因此，我们可以通过产生分歧的控制多通道舵在分歧的时间在一个信号周期的输出脉冲。下面按时器溢出间断产生PWM6标的目的舵控制(此系统只使用2个标的目的舵)同时进行。我们想控制信号输出端口6标的目的舵分别端口C.0 portc.5。高的时候电平分别是的Th0 Th5的。控制步骤如下：拉端口C.0第一，按时器按时TH0。当按时器溢出引起间断，端口C.0上拉下来，而portc.1拉按时器时间Th1.Do

45、这个过程中，一次又一次地直到portc.5拉低。然后按时器时间20毫秒(TH0+TH1+ TH2+ Th3的+ Th4为+ TH5)，并完成一个信号周期时序。标的目的舵控制值计算的流程图轨范和标的目的舵控制轨范图。图2和图3所示。图1.直流电动机的控制流程图。图2舵控制值计算轨范流程图图3舵控制轨范流程图四路径规划 无论机器人能够按照预先设定的路线走，关键的是，机器人可以采取直线和改变标的目的准确地按照控制的要求，并继续转动标的目的后沿直线行走。如果没有适当的算法，机器人会走明显S-形曲线，不仅影响速度，但也很容易偏离领先的线。机器人运动分为以下操作按次：寻找信号线，直来直去，计算的数量网格位

46、置，并改变标的目的。A.线追踪本设计采用三个传感器S1，S2，S3，以确保直来直去的机器人。这三个传感器分布在前面的机器人的三角形图案。当机器人移动前进，其前进的状态，决心通过这三个传感器的传感器的反馈值。当机器人有偏转，电机的速度指令进行的，以双方的电动机产生的速度差，并导致机器人回头向相反的标的目的引导网上。如机器人的速度快，这是很容易泛起的三个从领先的线，这是无法一起传感器偏离进行调整。因此，我们设计状态存储变量s_flag记录调整过程。当机器人偏离了领先行的左边，MCU作出命令的速度差电机控制模块，使机器人移回的右侧，并分派一个值，以上面的状态变量同一时间(例如，1)。类似地，当机器人

47、在偏离标的目的相反，将实施适当的行动，除了分派的另一个值(例如0)的状态变量以示区别。在调整过程中，机器人偏离必然的角度，因为调整滞后，虽然以电机。正如前面提到的外，所有三个传感器偏离领先行。然后，机器人会做出更大的调整的速度差异基于状态变量。调整的过程是，机器人将被作出了类似的行动，除了增加加快调整差别(完全停止电机)机器人重返领先行，当状态变量如图1所示，这表白，该机器人已作出调整有关在它的左边的偏差，但速度差异是不够的使机器人回到赛道。当状态变量0，调整过程可以类推。明细行跟踪处理流程示于图中.4。图4在线跟踪流程图B.行计数机器人的定位是通过计算行，即通过记录十字交叉口完成号机器人通过

48、。这个过程也被实现传感器。计数的线阵传感器安装在机器人的远处侧的轴，防止直引出线由于频繁的调整干扰计数线传感器机器人偏转的原因。在这个过程中，算法的选择是也是非常重要的。由于信号线的宽度达到30mm和扫描是一种快速的，它有可能反复计数，如果按时器每次计数传感器扫描黑线。传感器值不会跳以30mm的信号线。如果我们采用的上升或下降沿计数，按时器计数只有当传感器值从0到1，从1到0跳，所以我们可以防止泛起迭代计数。线的流程图计数示于图.5。图5行计数流程图C.旋转控制通过设置在两台电机指定机器人的转折点同样的速度，使机器人围绕斧中点两个电动机。机器人的转动标的目的和速度是由电动机。转向角的控制实现由

49、前端的计数线传感器。以90度的大转弯例如。当机器人开始转动时，它使扫描前端在同一传感器。当传感器扫描信号线，这暗示转折是在正确的地方，马达停止转动。至于转弯是否到位决意由前端传感器是否已检测到的信号线，时间开始扫描也是非常重要的。如果传感器是以上的信号线，在开始扫描时，机器人会不转。因此，它应该被延迟一段时间后再扫描。这是更好地延误，直至电机转动的一半角度。需要测试的具体时间按照速度。当的转动完成时，机器人的姿势有一些偏转由于惯性。当机器人行走直再次，所需的调整是比力大的，所以它是必要时作出轻微调整。调整实际上线追踪过程中低转速下。的流程图的转动控制示于图中.6和主流程图示于图.7。五结论按照2008年比赛要求和轨则亚太机器人大赛，控制软件本文设计的竞赛机器人。在控制系统，AVR单片机芯片的计算机语言和C语言用于实现直流电机的控制，舵伺服控制，并机器