Getting more self-driving cars on the road could spell the end of potholes
Car makers and AI experts across the globe are advancing technology to get fully autonomous cars onto our roads. Yet these technologies are having surprising knock-on effects elsewhere.
The mapping systems of Google’s Waymo, Tesla, BMW (and more) can also improve the surface of roads. This is because LiDAR, the laser system that acts as the “eyes” for autonomous cars, can identify divots, dips, cracks and signs of stress in the road ahead. In the UK, potholes are estimated to cause as many as 1 in 10 mechanical failures and cost drivers £730 million every year.
This same mapping technology could also be used by railway engineers to look for cracks in tunnels, or spot weak points in bridges.
More pertinently, technology exists that could automatically send reports of these potholes and weak spots to the relevant local authority so they can be filled in or strengthened before they cause any damage to cars.
LIDAR vs potholes
LIDAR bounces light off objects to see where they are, in a similar way that radar uses radio waves and sonar uses sound. LIDAR systems emit pulses of light outside of the visible spectrum and record how long it takes for them to bounce back. The time it takes for the light to reflect back tells the sensor how far away it is, and reveals its shape.
The direction and distance between the sweeping LIDAR sensor and each detected object are plotted on a 3D map called a “point cloud” in order to build up a detailed picture of the surroundings. While radar works well over long distances, and sonar is better suited to short distances, LIDAR provides a middle ground.
Earlier this year, Mitsubishi Electric revealed its Mobile Mapping System (MMS). It uses the regular LIDAR-based cameras and sensors to create “highly precise dynamic 3D maps”. The system, which can be fitted onto any family-sized car, creates these “clouds” of laser points as the lasers hit objects and send back reflections and shadows with a precision of 10cm or less. This laser cloud additionally uses what resembles heat mapping technology to spot changes in road height using a colour scale.
READ NEXT: What is LIDAR?
MMS takes things a step further by using “different extraction technology” to specifically spot and flag changes in the landscape. This could include larger changes such as roadworks or new signage and lane markings, or could identify more subtle changes in the road surface. Rather than constantly feeding full maps containing reams of data back into the system, which is both costly and time consuming, MMS only sends details of the changes, significantly speeding things up.
During a drive around Tokyo, we recently got to see this technology in action. As we turned a corner, an area of red appeared on the left of the display to highlight a drop in height in that region. Shortly after the car swayed and dropped into this dip. This was originally designed to help cars plot the terrain and adjust their speed or road position in response, yet could be extended to road maintenance.
What’s more, by creating dynamic maps in conjunction with ME’s on-board AI, this system has the potential to alert local authorities to changes in the road surface in real time. A pothole could be made, spotted and flagged, and filled within hours