Volvo Thinks Magnetic Roads Will Guide Tomorrow’s Autonomous Cars – Wired

Image: VolvoImage: Volvo

Image: Volvo

We live in a world where satellites and microprocessors can tell us how to get anywhere we want to go, but Volvo believes the future of transportation lies with something much simpler: magnets.

Yes, magnets. Volvo argues that magnets, unlike electronic transmission, are unfazed by poor weather conditions or obstacles in the road, so they can reliably guide vehicles along the road. It recently completed a research project to test the theory, using a road embedded with magnets and a vehicle outfitted with specialized sensors to determine how accurately and reliably the car could discern its position.

Volvo tested its theory at its test grounds in Hällered, Sweden. One of the biggest challenges was developing sensors capable of receiving data while speeding over a small magnet. Engineers calculated that a car traveling just over 90 mph would need a sampling rate of at least 400 readings per second. Typical magnetic sensors can only handle about three samples per second and need to be within centimeters of the magnet they’re detecting.

So Volvo built a rig with five sensor modules made up of 15 total smaller modules of Honeywell HMC1053 magnetoresistive sensors, all of which sent information to a single circuit board. The setup was good for up to 500 readings per second, making it fast enough to handle high-speed driving.

Volvo first tested the sensor system, mounted to an S60, on a “forest road consisting of a layer of stones of various sizes and on top of that a thick layer of soil had formed over the years,” according to a report on the project. Engineers lined the road with neodymium magnets 20 mm in diameter and 10 mm thick, and ferrite magnets 30 mm in diameter and 5 mm thick. They were placed in plastic tubes buried vertically in the road, with the mouth of the tube just below the surface. The first run with 100 magnets along 100 meters of road proved the system, when coupled with data collected from the car (speed, for example), could calculate the car’s position to within 10 cm when traveling just under 45 mph.

Further testing on a paved roadway, onto which engineers glued the magnets to the asphalt. The results were similar, but more compelling because the installation was far easier–and could be applied to existing roads. The sensors calculated the car’s position to the same margin, at speeds exceeding 90 mph.

Volvo tested their sensor system at speeds of up to 90 mph.Volvo tested their sensor system at speeds of up to 90 mph.

Volvo tested their sensor system at speeds of up to 90 mph.

Despite the positive results, Volvo realizes the idea sounds a bit preposterous. So it included some calculations showing the economic feasibility of this kind of system. The sensor rig, if made to a scale of 50,000 units, would cost about $109 per unit. The infrastructure overhaul for a major highway comes to about $24,405 per kilometer. That buys you seven parallel lines of magnets spaced 2.8 meters apart along a typical two-lane road.

For comparison, they used data from the California PATH project involving self-driving buses that navigated with, yes, magnets. Volvo used the PATH estimated cost of about $22,179 per kilometer.

Volvo’s long been interested in autonomous technology to improve safety. It started testing autonomous cars in 2011, and hopes to have self-driving cars capable of reaching 30 mph before long. It also has unveiled the Drive Me program, which promises to have 100 autonomous cars on the roads of Gothenburg by 2017.

The Swedish Transport Administration (Trafikverket, as it’s known), partially financed the magnet research.


Write a Reply or Comment:

Your email address will not be published.*