(Yicai Global) March 22 -- Driverless cars are providing many opportunities to advance technology. In order for them to become a reality, the technology must enable a vehicle to navigate through traffic with no human interaction. Research into new imaging technology has created promising avenues.
With major companies like Baidu testing cars in Beijing and many companies in the Guangzhou region, China will see more driverless cars on the road in the near future. Many Chinese companies also work with California companies in the U.S. to develop and refine technology. With more Chinese government-improved tests and the high interest of Chinese citizens, China will remain at the forefront of driverless technology and adoption.
In a fascinating development, researchers at Stanford University have created a new laser-based technology that can effectively detect objects that are out of the direct line of sight. In other words, the system can see behind things and around corners.
How Laser-Based Technology Works
The process works by collecting data from scattered light particles and reconstructing them into a three-dimensional image. For example, light from a laser pulse is bounced off a wall and the resulting light particles reflect off unseen objects, including around corners.
The system tracks the path of those particles and reconstructs the shape of any hidden object. Sophisticated algorithms also measure the time it takes for the light particles to bounce back to the system sensors.
This technology, called non-line-of-sight-imaging, is an advancement from current laser-based systems. Established imaging guidance systems use light detection and ranging (LIDAR).
LIDAR works in a similar manner, by measuring the distance from the system to a target. By aggregating all the measurements, LIDAR can create a three-dimensional map of the area around a car. LIDAR, however, only works with visible objects in the vehicles route. The recorded pulses are linear, meaning they only measure straight paths.
The non-line-of-sight-imaging technology collects more than straight pulses. Light can scatter in several different directions and by measuring those pulses as well, a larger field can be mapped. This has the potential to help driverless cars avoid dangerous blind turn collisions, adding to their expected decrease in car accident injury rates.
Advances in Technology Will Increase Safety and Offer More Opportunities For Use
LIDAR is currently the most promising form of vehicle guidance system. It has already been used for aerial mapping, agricultural studies, and archeological research. The new non-line-of-sight system has the potential to refine these uses. For example, if someone is trapped beneath rubble, the new system may be able to pinpoint their exact location.
Matthew O’Toole, co-lead author of the research paper, thinks current LIDAR technology can be used to include non-line-of-sight imaging: “We believe the computation algorithm is already ready for LIDAR systems.The key question is if the current hardware of LIDAR systems supports this type of imaging.”
Driverless cars are the wave of the future. The first cars could be available for consumers as early as 2019. Many market watchers believe we will see a broad consumer mass market within the next decade.
In China, testing approval is handled on a city-by-city basis, with Beijing and Shanghai showing much interest. Government regulations are now in place for car makers to proceed. In exciting news, Baidu CEO Robin Li recently said he believes driverless will be on the road in China in the next three to five years, much sooner than in the US.
The key to success, of course, is safety. With LIDAR and the new non-line-of-sight technology creating more detailed maps for cars to navigate through, the future may be here sooner than expected.
[Disclaimer: Yicai Global is committed to providing an open forum to air a diverse range of views. The opinions expressed herein are the author’s alone. Yicai Global has redacted this article to conform to our style and usage guidelines, but neither validates its factual nor endorses its editorial content.]