The name of our project is the instrument bike. The main function of our project is to collect some information about the bicycle, such as the speed position of the road, by collecting, storing and analyzing the data. Initially, we placed the data collector (Microcontroller and Accelerometer) on the bicycle, and then the data can be uploaded to the cloud server and mobile APP via the wireless network, and then the data is analyzed by MATLAB to upload the data to the website.

Instrumented bike is the future of Road Condition Assessment

This data collector consists of two accelerometers, a GPS, a power supply, an openlog, a particle photon, and an Arduino Nodemcu. Accelerometer and GPS data is stored to the SD card by particle photon and Arduino Nodemcu and uploaded to the mobile app and cloud(firebase). The SD card stores the accelerometer data for the corresponding time, that is, the x, y, and z-axis values and the GPS data. When the Z-axis data of the accelerometer is greater than 0.95, we judge that the bump will be generated, and the synchronization is displayed on the mobile phone and the cloud.

Our data processing software is MATLAB. The MATLAB classification algorithm is based on the data obtained from the accelerometer for road condition determination. Our classification algorithm is based on the application of neural network algorithms. According to our plan, our raw data is the acceleration values measured in three directions by the accelerometer. We divide eleven bump data into one sample and perform averaging, difference or Fourier transform on each sample to generate training data for the neural network algorithm. Then, the neural network algorithm will train through the training data. Finally, the training data is used as the test data to calculate the correct rate of the algorithm to judge the reliability of the algorithm.

For this part, it mainly includes five functions--scan the QR code, collect data from hardware and update, bump warning, line chart display and map display.Then, we plan to data collection and Interact with firebase as a cloud. After that, we plan to realize the bump warning, which contains mobile and wear warning via vibration when a bump is approaching.