Autonomous Driverless Car Application - BlueBird Electric

Task:

‘BlueBird Electric` is an electronic company that develops diverse types of innovative autonomous robotic units for different industries to support a wide range of systems.

One of the future objectives of BlueBird Electric is to be a leading company on autonomous technologies, particularly on autonomous driverless cars. You have been asked as an engineer in the ‘BlueBird Electric’ company to develop an autonomous driverless car application. You can use any programming language you are comfortable with. In this manner:

1 Comprehensive understanding of the scientific principles and concepts relevant to and Intelligent Machines engineering

2 Awareness of relevant regulatory requirements governing engineering activities in the context of Intelligent Machines engineering

3 Apply and investigate new and emerging technologies

4 Plan self-learning and improve performance, as the foundation for lifelong learning

Provide a detailed, professional report for your company that contains the following:

1- A review/insight of

i) Current autonomous driverless car applications along with their main components employed by three automobile companies,

ii) Why and how Deep Learning (DL) and Reinforcement Learning (RL) techniques are recently being forged together to result in better Artificial Intelligence (AI) applications by analysing successful real-world application,

iii) Likely use of this forging approach in autonomous driverless cars within smart transportation and smart city,

iv) How 5G communication technologies will revolutionise the autonomous driverless cars within smart transportation and smart city.

2- Development of a DL application that satisfies/includes the following features:

1. Build a faster R-CNN (for specifying detected objects with bounding boxes),
2. The software algorithm should be trained with car bounding box. Detection accuracy is expected to be greater than 95% regarding sensitivity and specificity values,
3. Use one camera as a sensor (computer vision),
4. Incorporate natural language processing (i.e., text-to-speech) into the application,
5. Provide driving information as a speech output along with the bounding box for the detected objects on camera images,
6. The software algorithm is required to detect road lanes drawn by white lines while performing other tasks, which requires a multi-threaded programming,
7. The software algorithm is required to measure the distance between the driverless car and the car in front and at left and right lanes with respect to the position of the camera while performing other tasks, which requires a multithreaded programming

3- Test the system using a video captured with driverless car unit and insure that this subsystem can be integrated into any autonomous driverless car unit; explanation with codes and proof of demonstration is required within the developed software regarding any operating system.

4- Proposal for Driverless Pod Sensor– Controller – Actuation System that could utilise the DL software algorithm developed above; Block diagrams are required.