| Summary: | The project calls for the application of engineering and programming knowledge to develop and implement an autonomous vehicle, the Nanyang Venture 11 (NV-11) which will be taking part in the Shell Eco-marathon Asia 2019. Throughout the project, different task had been allocated to collaborate as a team to build the NV-11. A 3-phase Brushless DC (BLDC) electric motor controller that was designed by the previous student had a few flaws which were the instantaneous current surge that cause the components on the circuitry board to be damaged. The implementation of the Soft Start code improved the performance and prevent it from any sudden current spikes occurring. The software kit by Renesas Electronics Corporation was used to program the microprocessor board to determine the clockwise and anti-clockwise rotation of the vehicle wheels. The NVIDIA Jetson TX2 is a supercomputer which is an embedded Artificial Intelligence computing device that acts as the brain on the autonomous vehicle NV-11. It operates using Ubuntu 16.04 Operating System and has the compatibility of using the Robot Operating System (ROS) as a platform that provides a set of software libraries and tools to build an application. A virtual environment was designed as a simulation to test the functionality of the autonomous vehicle using the data from the exterior sensors like the Lidar sensors, Zed camera, and Radar sensors. Using TensorFlow which is an open-source software library for machine learning to gather data for obstacles avoidance and detection. The Tinkerforge IMU Brick 2.0 is a device which is equipped with a 3-axis accelerometer, magnetometer (compass) and gyroscope. It provides the quaternions data which is a mathematical notation for representing orientations and rotations of the vehicle in three dimensions. It was implemented as part of the simulation to determine the exact location in the real world, comparing with the virtual world.
|
|---|