Roadside sensors and ML would warn German drivers of wildlife hazards

Radar and other sensors connected to a machine learning system could someday help German drivers avoid collisions with deer. (Spectrum Instrumentation)

Two German universities and tech partners are designing a system of sensors along roadways paired with machine learning to spot deer and other wildlife to then warn drivers in an attempt to lower accidents.

The so-called SALUS system is envisioned to be spread across Germany. It will rely on radar, optical cameras and infrared sensors to spot animal movement that are installed in posts 40 kilometers apart. That data will then be automatically interpreted to predict the behavior of wild animals and separate them from cars and motorcycles. 

“We are at a critical part of the project which is the classification of detected objects, which has never been done before,” said project leader Prof. Hubert Mantz from Ulm University of Applied Sciences. Heilbronn University is also a partner.

Fierce AI Week

Register today for Fierce AI Week - a free virtual event | August 10-12

Advances in AI and Machine Learning are adding an unprecedented level of intelligence to everything through capabilities such as speech processing and image & facial recognition. An essential event for design engineers and AI professionals, Engineering AI sessions during Fierce AI Week explore some of the most innovative real-world applications today, the technological advances that are accelerating adoption of AI and Machine Learning, and what the future holds for this game-changing technology.

Mantz said the roadside approach makes sense because it will take years to for driver assistance systems with animal alert technology to be spread across various car models at all price points.  Also motorcyles won’t have the same functionality of car alert systems.  With roadside technology, drivers without warning systems might be alerted to hazards with road lights and digital signs. 

In addition to measuring data from radar, optical cameras and infrared cameras, additional sensors could measure pollution.  Because the system needs to be spread across Germany, it needs to be inexpensive and solar-powered to reach areas outside the electrical grid.  The communication system between the units will be based on a low cost, low power Long Range Wide Area Network running on an unlicensed frequency that can reach the necessary 40 kilometers between detection posts.

Spectrum Instrumentation of Hamburg is providing a digitizer card to collect the micro-Doppler radar data. The card starts at $3,680 with an unusual five-year warranty that will help lower maintenance and repair costs as the detection units proliferate, Mantz said. The Spectrum digitizer runs on four channels and 10 MHz bandwidth to allow processing of data simultaneously in real time.

RELATED: Collision avoidance sensors market to grow at 21.45% CAGR through 2026

Suggested Articles

Power management is critical for many products. One expert advises pushing complexity to areas where energy is less of a concern.

HP leads the pack, but Apple sees 36% surge in notebooks, desktops compared to a year ago

New York City residents moving upstate will give hundreds of towns extra tax revenue to invest in new streetlights with IoT sensors, NYPA foresees