Table of contents
3. Summary of the reasons for the competition
2. Teammates do not cooperate well
3. There is a problem with the overall process arrangement
I. Introduction
This year, I participated in the Provincial Electronics Design Contest, and I did the control car topic. The test has been completed this afternoon. Although the result was not good, I have been reviewing and summarizing after the game, so I specially wrote this article to record it. (This is the first time I write a blog, please criticize and correct the bad and wrong places)
2. E-sports review
This is my first e-sports competition, and there is a high probability that it will be my last e-sports competition during college. From the preparations on July 2, 2022 to the start of the competition on July 27, 2022, and the end at 8:00 p.m. on July 30, although I didn't try my best, I basically tried my best. In the process of preparing for the electric competition, I learned the basic communication protocol of STM32 single chip microcomputer, such as clock, IO configuration, serial communication, external interrupt and IIC, and the basic control method of smart car, such as differential speed turning and duty cycle control. Motor speed and PID closed-loop control methods. In addition, I also learned some necessary peripherals, such as MPU6050 gyroscope, HC-05 Bluetooth module, encoder motor speed measurement, infrared tracking module, voltage regulator module, motor drive module L298N. In the end, I also briefly learned about the MSP430F5529 single-chip microcomputer (I didn't know at the time that the problem of the car restricted the TI chip, but now I think of it, I should directly use the TI chip). During the four days and three nights of the electric competition, when we saw the problem, we thought the problem was not difficult, and soon we confirmed that the second car used ultrasonic ranging to achieve tracking. Both cars used infrared tubes for tracking, communication This is achieved using the interconnection of two Bluetooth modules. This plan is feasible, but it is not smooth in the process of doing it. Since MSP430 is not used much, basically all configurations have to be configured and debugged on site, which wastes a lot of time, which is a big reason for the bad results of a competition. For example, we encountered many problems when configuring the clock of MSP430. At that time, we did not understand the USC clock configuration of MSP430, and spent a lot of time learning. It took more than three hours to configure the communication between 430 and OPENMV, but it was solved by changing the baud rate to 115200. As for why, it is still unclear. I have been configuring these peripherals for the first two days, and I started to tune the car until the next night, but I also encountered many problems during the tuning process. The biggest problem is tracking, and even failed to track call out. In summary, there are many problems in this electric competition. It seems that everything has encountered problems. Fortunately, although it took a lot of time, every peripheral was successfully configured in the end. The most regrettable thing is that the tracking has not been done. If the tracking of the two cars is done, maybe the result will be better. I record the failure reasons and solutions below, hoping to provide experience and lessons for future e-sports players, and record the results of my own trial and error.
3. Summary of the reasons for the competition
1. Poor preparation
I only briefly studied the MSP430 chip. If it is well prepared, many peripherals can be used directly, so that more time can be left for debugging the car. Too many questions and blow self-confidence.
2. Teammates do not cooperate well
The division of labor of our teammates is a software, a hardware, and a combination of both. This is a common division of labor in the group, but there are many problems when we subdivide the learning tasks. The other two teammates are basically learning the open mv camera and the K210 camera, but the car problem basically does not use the camera (the camera can be used to track and identify , but the effect was not good after we tried it), so basically I was doing it alone, and the two teammates often couldn't help if they wanted to help. Unless you are led by a great master, it is recommended that teammates do not divide the learning tasks too clearly, and try to squeeze out more time to learn more modules. In addition to the lack of cooperation in learning tasks, the tacit understanding between teammates is not enough, there is not much communication, and sometimes the team members cannot communicate with each other, which greatly affects efficiency and morale. It is suggested that teammates should take the initiative to talk about what they bought, what they need, and what new directions they have. They cannot be alone, and three people must unite to solve more problems.
3. There is a problem with the overall process arrangement
When doing the small car problem, we configure the basic peripherals first and then start to do it. We thought that after configuring the peripherals, we can solve the problem smoothly. In fact, many things seem simple and we start to do it. There will be a lot of problems, so that we didn't even use the PID algorithm and encoder speed measurement in the end. We should do the topic first, and configure what modules will be used in the process of doing it. When debugging open-loop or closed-loop, it is recommended to use the open-loop to realize the tracking first, and then use the PID algorithm to optimize smoothness and speed control. .
4. Experiences and lessons of the small car problem
- Make the hardware well, including where the tracking module is located, what motor to choose (a good motor is really easy), what kind of wheels to use (swivel wheels or bull’s-eye wheels) and frame (whether it is convenient to install peripherals) And the center of gravity is stable), the hardware can greatly reduce the complexity of the program
- For four-wheel drive, two-wheel drive, and several-wheel vehicles, you should first learn two-wheel drive and three-wheel drive, and then if you have enough preparation time for two-wheel drive, try to learn from four-wheel drive. Four-wheel drive cars have higher stability and more room for adjustment. Also a lot.
- Don't underestimate any function, the difficulty of the game is likely to be the point that is usually overlooked the most
- For each function, learn more solutions, such as infrared, grayscale and camera for tracking, so that you can change the solution when you encounter a solution that cannot be realized during the game
- Record each debugging data, which can save a lot of debugging time
5. Summary and harvest of the e-sports competition
- Prepare more, don't take tricks, the car question is very flexible, you must learn the basics to deal with it easily
- Communicate more with the bosses, and stand on the shoulders of giants to see the world. These bosses include both school bosses and online bosses. Communicating with them can save a lot of detours and improve awareness
- Diligence and hard work are always the main theme. If you don't study hard, you can't get good results.
- Don't be afraid of difficulties, although every difficulty is annoying, but you will be very happy after solving the problem
- Summarize more, accumulate more, learn more, don't think about achieving it overnight, and take every step down-to-earth is the most realistic