In the coming weeks, we will be discussing different aspects of autonomous vehicles in this blog. Everyone is familiar, at least at a high level, with the concept of an autonomous vehicle or in other words a vehicle that can drive itself. However, the complexities to make this happen are amongst the biggest challenges in engineering today. Imagine all the different scenarios and corner cases that we humans deal with today when we drive. We must negotiate rush hour traffic, anticipate a child running in front of our car, weave around debris in the road, merge into high-speed traffic, react to traffic signals, follow detour signs and so much more. These do not seem like a big deal to humans because we have a highly developed brain, but how do we train a computer or series of computers to perform these tasks without failure?
The answer has many facets that we will explore, including how modern technology such as cameras, FPGAs, lidar and artificial intelligence are used and what does “autonomous” really mean? We will begin the exploration in the next post by discussion the 5 levels of autonomy as defined by the Society of Automotive Engineers (SAE).