When we last visited Stanford to take a look at the Volkswagen Automotive Innovation Lab (VAIL) work on robotic vehicles, we had the chance to sit in the Junior 3 autonomous car while it slowly backed itself into a parking space. Count us out for a repeat performance with the original Junior autonomous car, which recently learned how to slide park--a process that involves putting itself into reverse, accelerating to 25 mph, quickly braking while simultaneously turning the wheel, and making a 180-degree slide into the parking spot. It's like an extreme form of parallel parking.
Popular Science explains:
Junior usually uses a model of the car's dynamics to control steering, gas, and braking. This is relatively easy when the car is moving straight ahead, but when it's sliding sideways across the pavement, the variables -- friction of the tires, texture of the pavement, grade of the roadway -- are far more diverse. That means working from a fixed model is far more difficult. So the team also implemented a two-prong approach. There's the physics-based simulator that understands how the car moves in space, and a second system that draws from past demonstrations of the car sliding across the pavement. That second system works more or less by trial and error; it draws on the cars "experience," repeating the control inputs it used in other successful slide maneuvers to try to reproduce the successful results.
In other words, the slide park is one step on the way in allowing Junior to think for itself and make the right choices in emergency situations. And while many drivers will be reluctant to step into a fully autonomous car, they will probably be more comfortable with a vehicle that can make split-second decisions in the moments before a crash.