Perception with Self-Driving Vehicles
Perception with Self-Driving Vehicles
Self-driving vehicles use a wide array of sensors to perceive their surroundings, on-board computers to combine and process the data from all of those sensors, and one or more motors to safely move the vehicle along the road using AI (artificial intelligence). Self-driving vehicles often make use of high resolution cameras to be able to recognize things such as pedestrians, road signs, and other vehicles. Vision data from the cameras is often combined with sensors that use lasers to measure distance between the vehicle and other objects. This allows the on-board computers to make decisions based on what types of objects are in their environment, and how far away each object is.
Motivate Discussion
Q: What do designers need to do to make sure that self-driving cars are safe?
A: Possible replies will include making sure that the car sensors provide sufficient and precise information, practice driving on different types of roads, practice driving in different types of weather conditions.
Q: Why is it important that the Vision Sensor used for the car will have to be able to practically tune itself?
A: The Vision Sensor will have to be able to detect objects regardless of the environment. This includes being in a tunnel, snow, rain, at night, direct sunlight, etc...
Q: How will using self-driving cars affect the environment?
A: Students may think of several pros and cons to this question. An example of a pro would be that self-driving cars will be safer, which should lead to fewer accidents. This will reduce the need to replace vehicles. A con would be an increase in the usage of certain resources, such as gasoline and petroleum, due to an increase in driving.