map

Toyota has developed a high-precision map generation system that will make use of data from on-board cameras and GPS devices installed in production vehicles.

Developed to aid the safe implementation of automated driving, Toyota’s new system will be exhibited at Consumer Electronics Show 2016 in Las Vegas, US, which will be held from 6 to 9 January.

This new system uses camera-equipped production vehicles to collect road images and vehicle positional information. This is subsequently transmitted to data centres where it gets automatically pieced together, corrected and updated in order to produce high-precision road maps.

It is necessary to have an understanding of road layouts and traffic rules for the successful implementation of automated driving technologies.

"It is necessary to have an understanding of road layouts and traffic rules for the successful implementation of automated driving technologies."

High-precision measurement of positional information also needs information on dividing lines, curbs, and other road characteristics.

Map data for automated driving purposes is often generated by utilising specially built vehicles equipped with three-dimensional laser scanners. The vehicles pass through urban areas and on highways, and then the data is collected and manually edited to include information such as dividing lines and road signs.

However, due to the infrequent nature of data collection, such maps are not usually updated, thereby restricting their usefulness.

On the other hand, Toyota claims that its newly developed system uses automated cloud-based spatial information generation technology, which has been developed by Toyota Central R&D Labs, to generate high-precision road image data from the databanks and GPS devices of designated user vehicles.

Although a system depending on cameras and GPS in this manner has a greater degree of error probability than a system using three-dimensional laser scanners, positional errors can be lessened using image-matching technologies that integrate and correct road image data collected from multiple vehicles, as well as high-precision trajectory estimation technologies.

Therefore, the system’s margin error gets restricted to a maximum of 5cm on straight roads.

By utilising production vehicles and existing infrastructure to collect information, this data can be updated in real time. Furthermore, the system can be implemented and scaled up at a relatively low cost, the company claimed.

In order to support the spread of automated driving technologies, Toyota intends to include this system as a core part in automated driving vehicles that will be made available in production vehicles by around 2020.

Image: It is necessary to have an understanding of road layouts and traffic rules for the successful implementation of automated driving technologies. Photo: courtesy of Toyota.