David Boian - PTU sensing module
Imran Tauqeer - PTU actuating module
Jack Thompson - LED & timer module
Jia Xu Chee - sequence module, minutes
Jiapeng Liu - LIDAR module
Prithik Saini - PTU actuating module, python game module
The code contains 4 self-containing stages that works separately. There's one stage each before and after the 4 stages that require coding.
Done by Jiapeng Liu.
The work of detecting distance with LIDAR mainly focuses on two aspects:
- Optimization of detection distance algorithm
- Visualization of detection distance
-
Optimization of detection distance algorithm LiDAR may make errors when detecting distance due to external environmental interference or target angle interference. 1. Data processing and filtering: Consider using multiple continuous measurement data for judgment and decision-making, rather than just a single data point. By statistically averaging multiple data, the accuracy of distance measurement can be improved. 2. Threshold setting: Adjust the threshold according to the detection distance requirements. Due to the excessive sensitivity of the LiDAR detector, the distance data will bounce back and forth in two areas, such as the too-far area and the detected target area. By adding a buffer space to solve the problem, leave around 5 buffer spaces at the threshold setting for each region. For example, the threshold for the target area is distance < 50, and the threshold for areas too far away is set to distance > 55.
-
Visualization of detection distance Arduino connects a 1.3-inch OLED screen to display the detection distance type (ok, too far, too close). The communication between STM32F3 Discover Board and Arduino uses a logic level detection method.
Connect all necessary equipment according to requirements and continuously change the distance between obstacles and LiDAR equipment for testing. When the distance is greater than 55, the OLED display screen displays "Too far". When the distance is less than 25, the OLED display screen displays "Too colse". When the distance is within the range of 30 and 50, the OLED display screen displays "Ok"
Done by David Boian, Imran Tauqeer and Prithik Saini.
- Check that each pin received when high.
- Observe servo response in response to high pin.
- Check pins enabled correctly by checking the data we received.
- Check if servo responded to pin data.
- Check if we are receiving data after sensors enabled.
- Perform math on raw data and observe response.
Done by Prithik Saini.
The python script runs simultaneously with the STM32 code. When the passing condition: "RUN_PYTHON\r\n" is sent to the PC terminal via USART1, the game play. The user will have to hit the floating balloon using the keyboard inputs. It follows with displaying a set of instructions for the next stage before closing.
Modular testing can be done in the Game folder. Several python libraries are required (can be found in the files) to run this module in particular.
- The game is run separately to make sure the required libraries are avaliable to run on the system.
- The STM32 code is then run to test the passing case "RUN_PYTHON" is output to the terminal.
- Finally, the scrpit is run simultaneously with the STM code to make sure the module works as intended.
Done by Jia Xu Chee and Jack Thompson.
Modular testing can be done in serial_module folder. Since this stage can be done solely in STM, the important parts of the code are in serial.c and sequence.c. To test the module, call the function LED_Stage() in main.c. The expected output of the stage is as the flowchart shown above, where the terminal will output "pass" when the player inputs correct sequence; "fail" for incorrect input; "Restarting this stage" when restarting the stage; and "You've passed this stage" when the plater inputs 4 consecutive correct sequence; "Time's up, restarting this stage" when no user input for 10s after the LED sequence.
- Stewart Worrall MTRX2700-2023 GitHub Repository https://github.com/stewart-worrall/MTRX2700-2023/tree/main