Data from vehicles traveling on multiple roads enables Tactile Mobility to create constantly updated maps that can be used by customers such as municipalities, fleet managers and insurers.
Autonomous vehicles are supposed to be the future of urban transportation, yet AVs still can't make decisions the way humans do.
Humans gain a wealth of information through the five senses — sight, touch, smell, sound and taste. Meanwhile, autonomous vehicles respond to just one sense: sight.
How will fully autonomous vehicles provide safer transportation than human-controlled vehicles when they lack the knowledge and experience derived from the four other human senses? A startup in Israel is getting closer to figuring it out — and automakers want in.
Tactile Mobility has developed a software stack that allows self-driving and smart vehicles to turn data from nonvisual sensors into information about the road surface and the vehicle itself in order to "feel" the road beneath the tires and respond appropriately to the road conditions ahead.
Nisenbaum: Talks with automakers
The result is smarter and safer urban transit, CEO Amit Nisenbaum told Automotive News.
The system uses embedded and cloud software.
Nisenbaum says the company's embedded software uses proprietary algorithms and artificial intelligence to generate real-time insights about vehicle-road dynamics.
"About 50 to 60 self-driving vehicles equipped with our embedded software will drive on the streets of a particular city and collect information from existing nonvisual sensors," Nisenbaum said.
The embedded software analyzes sensor data such as wheel spin, wheel angle, rotations per minute and gear positions.
The results of all this data-crunching are fed back into the vehicle computer, providing better context for driving decisions.
"Vehicles embedded with this software are provided with insight on the road ahead, improving safety and user experience reaction time," Nisenbaum said.
Meanwhile, Tactile's cloud software aggregates data gathered by multiple vehicles on multiple roads to create constantly updated maps of road conditions and hazards, including bumps, cracks and potholes. This crowdsourced SurfaceDNA 3D model can be used by customers such as municipalities, fleet managers and insurers.
"Imagine the thought of taking multiple trips of different vehicles on the same road and overlaying the signals collected from each vehicle," Nisenbaum added. "Data collected are processed into machine-learning algorithms that tell a narrative about the road, and with that, we are able to create maps of the roads."
Tactile Mobility launched its mapping service in Haifa, Israel, in September 2018. Nisenbaum said that in addition to Israel, the company has implemented pilot programs for its technology in the United Kingdom and in an undisclosed U.S. city in the Midwest. The company is about to start another pilot program in Asia, he said.
Tactile Mobility is working with six automakers in Europe and North America, including Ford Motor Co., said Nisenbaum, who spoke to Automotive News in July. He also said the company was in discussions with several automakers to embed its software in mass-market production vehicles.
Send us a letter
Have an opinion about this story? Click here to submit a Letter to the Editor, and we may publish it in print.
