 |
19 Jun 2024
Aeva's sensor technology selected for federally funded initiative also featuring Siemens and Bosch.
Aeva, the Silicon Valley company developing advanced optical sensors for transportation applications, says it has been selected to take part in a German project to deliver an automated driverless train.
The Mountain View firm’s frequency-modulated continuous wave (FMCW) technology, which unlike conventional lidar sensors is based around chip-scale silicon photonics, will work alongside ten partners - including Siemens Mobility and Bosch Engineering - as part of the “AutomatedTrain” consortium.
Led by a subsidiary of national operator Deutsche Bahn and funded by the German Federal Ministry of Economic Affairs and Climate Action under its Digitale Schiene Deutschland effort, the project aims to create and trial driverless, fully automated passenger train dispatch and parking functionality by 2026.
The effort will see two regional trains equipped with smart sensors, including lidar, to enable full automation and reaction to obstacles, the same principle behind self-driving cars.
FMCW advantages
While it requires much more complex engineering and fabrication than conventional time-of-flight lidar technology, the FMCW approach offers advantages in terms of both detection capability and miniaturization.
Because it is based around a frequency-chirped laser source, FMCW is able to deduce both the position and directional velocity of surrounding objects - yielding an obvious advantage in automated driving.
But the technology is at a much earlier stage of development than time-of-flight, and analysts at Yole Intelligence do not expect it to have any impact on the automotive sector until 2028. Earlier applications are anticipated in industrial settings, however.
Christian Hauswald, the technical project lead of AutomatedTrain at Digitale Schiene Deutschland, said: “Aeva’s high-resolution lidar sensor combined with a compact form factor and advanced technology readiness level were in combination the reason for us to decide using Aeva for the AutomatedTrain project.
“We are using the best available technologies and are excited to have Aeva as a supplier in this project because they share our vision for fully automated trains.”
Aeva’s senior manager of business development, Jakub Zimny, added: “Digitale Schiene Deutschland and its partners are undertaking massive transformation to modernize the rail industry and will be setting the standard for automated train operations across Europe.
“We are excited to collaborate with them and provide next-generation sensing and perception technologies to create a safe and reliable platform that can scale to meet the needs of the entire industry.”
Shunting applications
According to the project brief, by 2026 a train from Siemens Mobility (Mireo) and a train from S-Bahn Stuttgart (Alstom BR 430) will be equipped with the relevant hardware and software.
"The Mireo train will cover the route from the siding area to the first station in the Wegberg-Wildenrath test center fully automatically and without a train driver," it states. "The train will brake automatically when obstacles are detected."
Aeva points out that the velocity detection capability of FMCW would allow trains to instantly see which objects close to tracks are moving and which are stationary, providing faster and more reliable identification of potential obstacles and hazards.
“In addition, Aeva’s unique localization capabilities allow for the real-time estimation of a train’s motion with six degrees of freedom, providing accurate positioning in areas where GPS signals may be unreliable, including tunnels and urban canyons created by tall buildings,” added the firm.
It is not the company’s first effort in automated trains - last year Aeva said that Railergy, a rail automation firm also based in Germany, had selected its FMCW technology to assist with shunting applications, where trains are separated, moved, and linked together.
And earlier this year Aeva revealed that its FMCW sensors would be widely deployed by automotive giant Daimler Truck. It is scheduled to begin adopting the technology in 2026, ahead of a production ramp to potentially hundreds of thousands of units in 2027.
 |  |  |  |  |  |  |
| © 2024 SPIE Europe |
|
