Computer-driven vans finish test trip of 8,000 miles from Europe to Asia

Four driverless electric vans successfully ended a 8,000-mile test drive from Italy to China yesterday.

The vehicles, equipped with solar-powered laser scanners and video cameras that work together to detect and avoid obstacles, are part of an experiment to try to improve road safety and advancing car technology.

The sensors enabled them to navigate through wide extremes in road, traffic and weather, to Shanghai while collecting data to be analysed for further research, in a study sponsored by the European Research Council.

Hide Ad
Hide Ad

"We didn't know the route, I mean what the roads would have been and if we would have found nice roads, traffic, lots of traffic, medium traffic, crazy drivers or regular drivers, so we encountered the lot," said Isabella Fredriga, a research engineer for the project.

Although the vans were driverless and mapless, they did carry researchers as passengers in case of emergencies.

They did have to intervene a few times – when the vehicles got caught in a Moscow traffic jam and to handle toll booths. And at one point a van stopped to give a hitchhiker a lift.

The project used no maps, often travelling through remote regions of Siberia and China. A computerised artificial vision system dubbed "gold", for generic obstacle and lane detector, analysed the information from the sensors and automatically adjusted the vehicles' speed and direction.

Hide Ad
Hide Ad

"This steering wheel is controlled by the PC. So the PC sends a command and the steering wheel moves and turns and we can follow the road, follow the curves and avoid obstacles with this," said Alberto Broggi of Vislab at the University of Parma in Italy, the lead researcher for the project.

"The idea here was to travel on a long route, on two different continents, in different states, different weather, different traffic conditions, different infrastructure. Then we can have some huge number of situations to test the system on," he said.

The technology will also be used to study ways to complement drivers' abilities.

Related topics: