Nissan racing to catch up with self driving cars
Tetsuya Iijima sits down behind the wheel of the Nissan Leaf and holds his hands just below the steering wheel, like hes about to juggle invisible apples. Its the obliged pose for engineers testing the prototype of Nissans self driving vehicles on the streets of Tokyowhere Nissan and other automakers expect their tech to drive customers inside four years.
Nissan was the 1st big automaker to promise production ready self driving cars by 2020although ones that will still require a drivers attention. Ahead of that goal, Nissans engineering teams in Japan and Silicon Valley discovered just how much learning they’ll have to give their machines in the syntax of driving.
This specific Leaf looks alike to the electric cars Nissan was selling for many years, but carries millions of dollars worth of extra hardware. Nissan equipped the car with twelve cameras and five radar sensors to offer a 360-degree field of mechanical vision. A laser rangefinder setup, or lidar, tracks close moving objects like pedestrians the unit in the Leaf is itself a prototype Nissan built with a supplier, intended to be more compact and cheaper to build with an eye on mass production .
And the hatch of the Leaf was filled with a mini server farm which can handle the data flow and make the tactical decisions needed to manage the car in traffic. Iijima estimates at its peak, the car will have to procedure three teraflops of data from its sensorsjust to manage the hustle of city driving, which Nissan considers the toughest challenge for a semi autonomous car.
In our 30-minute ride through central Tokyo, the Leaf makes some number of key choices properly. When a trash truck aggressively merges from the left at speed, the Leaf maintains enough space with the car in front to let the truck in. When traffic slows, the prototype turns on its blinker and changes lanes, then moves itself back. And thanks to a mix of GPS navigation and its cameras, the Leaf can read not just speed limitations but stop lights, properly halting itself during.
But its not a flawless performance. After one stop light, halfway through a left turn, the Leaf straightens out, aiming itself for a set of concrete pylons in a pedestrian walkway. Iijima grips the wheel, taps the brakes and turns the car more sharply, keeping away from any contact. The error, he says, probably came from the cars misunderstanding over where the curb ends.
We still have to train the software, develop the software to understand all the different situations, Iijima says.
ts the non verbal communications among drivers and pedestrians thats showing the toughest challenge for engineers. If you come to an odd three way stop with a blinking red light, you may be able to probably figure out quickly whos supposed to do what depending on other vehicles, a trick that self driving cars havent mastered still. The sensors in the Leaf that track moving objects rely on motion, a pedestrian whos standing still at a crossing may get overlooked by software.