Implementation basics for autonomous driving vehicles

By Jan Pantzar (VSORA) and Lauro Rizzatti

The automotive industry is delivering the first implementations of advanced driver-assistance systems (ADAS) for Level 2 (foot off the gas or break) and Level 3 (hands off the wheel) vehicles. Though it’s struggling to develop an autonomous driving (AD) system from L4 (eyes off the road) to L5 (completely self-driving and autonomous) vehicles. The challenge is turning out to be more difficult than anticipated a few years ago.

Implementing an AD system comes down to safely moving a vehicle from point A to point B without human assistance. This can be accomplished by a three-stage state machine called driving control loop that includes perception, motion planning, and motion execution. Perception learns and understands the driving environment, as well as the vehicle position or its localization on a map. The perception stage feeds environment and localization data to the motion or path planning that calculates the trajectory of the vehicle, in turn performed by the motion execution. If perception generates inaccurate data, the trajectory is going to be flawed. In the worst-case, it leads to catastrophic results.

A successful AD system implementation rests on a state-machine architecture that can formulate a truthful understanding of the environment, produce an efficient motion plan, and flawlessly perform its execution.

To read the full article, click here