Scenario. Using Big Data And Artificial Intelligence To Create Autonomous Cars

BMW clearly believes the cars of tomorrow will drive themselves, instead of relying on humans being at the wheel. Its not alone in this belief all the major car producers are investing in driverless technology, but BMW is more confident than most. The company has said it plans for its cars to deliver full level 5 autonomy by 2021.

Getting to fully level 5 autonomous cars relies on computer vision technology, which means teaching computers and machines to see like us humans do, but using cameras instead of eyes, and to interpret what they are seeing in a way thats similar to the human brain. Its essentially a very advanced type of image recognition technology the tech that powers a simple Google image search where machines are taught to sift through and categorise images, becoming better at it over time as they process more and more images. Only, in this case, instead of sorting through millions of Internet images of cats and dogs, computer vision in an autonomous vehicle means ingesting all of the input data from the cars sensors and cameras, and analysing that data on the fly in real time quickly enough, say, to perform an emergency stop if needed.

Questions:

1. Self-driving cars are a very popular concept now. To guarantee the self-driving cars can react quickly enough to emergencies, the AI system deployed on the car is required to make decisions at real time. What technology mentioned in our course is a must for the AI system of self-driving cars? Please justify your answer. (5 marks)

2. In regard to the technology you proposed in your answer to question 1, explain how it collaborates with the AI system and the sensing system of the self-driving car to navigate the car to run. (5 marks)

3. Before the AI system makes a decision of stopping the car, it needs to classify the current scenario using a classification algorithm. Suppose we use the following table as a training data set for the classification algorithm, please create the corresponding decision tree. You can either draw it using drawing software (Viso, Powerpoint, MS Paint, etc.) or on paper and take a picture of it, and upload the file or your picture using the submission link. (5 marks)

	Object	Distance (M=meter)	Speed	Operation
1	Stop Sign	<10M	>=20Km/h	Slow down
2	Stop Sign	<10M	>=40Km/h	Stop
3	Stop Sign	10-20M	>=40Km/h	Slow down
4	Stop Sign	10-20M	>=60Km/h	Stop
5	Human	<15M	>=20Km/h	Slow down
6	Human	<15M	>=40Km/h	Stop
7	Human	15-30M	>=40Km/h	Slow down
8	Human	15-30M	>=60Km/h	Stop
9	Human	<5M	>=10Km/h	Stop
10	Stop Sign	<5M	>=10Km/h	Stop