Project ID: AEE-2018-30
Students: Lucas Bondén, Ville Pirsto, Niko Luostarinen, Samuli Sirniö, Marius Baranauskas
Project manager: Lucas Bondén
Instructor: Victor Mukherjee
Other advisors: Vesa Korhonen
Starting date: 5.1.2018
The main objective of this project has been to design and build an electric bicycle conversion kit. Additionally, an AI implementation is to be developed for the conversion kit to provide assistance with the driving.
The project has started with a planning phase and subsequent division of the work into different sub-parts. These sub-parts are motor modelling and design, controller implementation, converter implementation, mechanical design and smart features with AI.
The work in each part has begun with theoretical learning, software learning and planning. Due to the very diverse work parts, the implementation and testing methods varies depending on the work. However, all work phases are a combination of software and hardware implementation.
A working kit with an AI assist reference implementation has been achieved. However, issues with the converter when powering it with a battery has caused complications in the project. Due to these issues the self-developed converter and motor controller have not been used in the kit. Thus, even if a working prototype has been completed, the project has not succeeded fully according to the project plan. Future improvements can be made by using a more powerful microcontroller, troubleshooting the converter and improving the mechanics.
All students have learned valuable technical and theoretical skills depending on which part of the project they have been working on. Furthermore, all students have learned about the ethics of working in a project, mainly teamwork and problem solving.
- Motor: The initial goal of building designing and building the motor was deemed to be unfeasible after doing a FEM-analysis of different motor geometries in COMSOL. The biggest problem was that materials that had a high enough permeability and were 3-D printable weren't possible to find. In the end a outer rotor 14-pole 12-tooth BLDC-motor was purchased to use in the bike.
- Controller: The implemented BLDC controller was devleeoped on an Arduino Uno. The developed controller works well in scalar control mode but there were complications when trying to use vector control of the motor because of limitations in the Arduinos performance.
- Converter: The converter consists of a three phase bridge implemented with MOSFETs and freewheeling diodes. The converter also has circuitry to power the two Arduinos and optical isolation between the power and signal side. Unfortunately the Converter was damaged when powered with the battery.
- Assist reference controller and machine learning app: The bike comes with an experimental machine learning application that utilizes Tensorflow machine learning library. The purpose of the app is to learn the rider's cycling behavior and automate the motor usage optimally for the cyclist's needs. First, cycling data is collected from the bike. The data is then fed to the app, which computes a neural network model. The model is then exported back to the assist reference controller, which uses it to predict the assist amount.
- Sensors: In order to make the bike follow the Finnish legislation and to implement machine learning sensor data was needed. The sensors measure pedalling speed, bike speed, acceleration and user speed reference.
- Mechanics: The motor can of the outer rotor motor drives directly on the back wheel by utilizing the friction between the tire and the motor. The motor stays disengaged from the wheel when not in use and swings against the tire when assistance is required.
Fig. 1. Close up of the kit