TEL AVIV, Israel — Israeli automotive technology startups hoping to mirror the success of Mobileye showcased their innovations for venture capitalists, automakers and Tier 1 suppliers here at the annual EcoMotion Week conference.
Jerusalem chip and sensor company Mobileye has grown to a stock market valuation of $33 billion because of its advanced driver-assist and autonomous driving systems business.
Automotive and mobility technology is a fast-growing Israeli industry as entrepreneurs bring expertise gained from developing missile defense systems and other military technologies, along with the know-how acquired in related disciplines to the sector.
There are more than 650 automotive and mobility technology startups in Israel, with 110 exhibiting at the conference, said Jennifer Schwarz, executive director for EcoMotion. More than 3,000 people attended Tuesday’s exhibition.
Israeli automotive technology startups have raised almost $200 million this year, according to IVC Data and Insights, an Israeli venture capital industry research organization. Investors poured more than $1 billion into Israeli automotive and mobility sector startups last year.
The companies at EcoMotion displayed a broad array of technology, much of it dealing with autonomous driving systems and electric vehicle components. Here is a sampling:
The Tel Aviv startup develops analytic tools to monitor scenarios encountered by advanced driver-assist and autonomous driving systems, including those edge cases that create the biggest safety risks.
Foretellix helps self-driving technology companies validate the data they cull from simulations and recordings of real-world test-car driving.
“Safety is the main barrier for mass deployment as the technology is very advanced,” said Gil Amid, Foretellix chief regulatory officer and co-founder. “But still, you’ll see reports once or twice a week on the failure of automated driving systems.”
Foretellix provides a big data analytics platform that aggregates testing results and looks for testing holes, telling users where more work is needed.
For example, an automaker might have to validate that it has tested a lane-keeping system that works from 0 to 100 mph. The Foretellix system analyzes whether enough testing was completed at every velocity point to ensure the automated driving feature’s safety, Amid said.
Foretellix raised $43 million in Series C funding this month, including investments by Woven Capital, the growth fund of Toyota and Nvidia. Volvo Group VC also is an investor.
NoTraffic develops technology that allows municipalities to adjust traffic signals to traffic conditions, meaning city traffic departments can set up rules for its algorithm to follow. If traffic, for example, reaches a specific density in one direction, the system can keep the green light open longer to help clear backups.
Tal Kreisler, CEO of the Tel Aviv company, said the algorithm can adjust signal patterns to change traffic when schools start or when there is heavy pedestrian or bicycle volume.
The NoTraffic system relies on a pod with a camera and radar mounted on the traffic light. It is capable of adjusting lights on the fly.
“The algorithm will predict if a vehicle is going to run a red light and delay the light turning green on the other side,” Kreisler said.
This Holon, Israel, startup is developing a solid-state lidar that would fit into current automotive styling. Its small size allows it to fit seamlessly into the vehicle, avoiding the fuel or energy penalties caused by the bubbles and cubes that house larger lidar with spinning components.
Opsys Technologies completed a $51.5 million Series C funding round this year. Hyundai is one of its investors.
The company said that it is making progress in shrinking the lidar so that it can be integrated into a windshield.
“It’s not on the roof, it works behind glass, and it hardly produces any heat,” Opsys Technologies CEO Rafi Harel said in a statement.
Automotive parts supplier Huayu Automotive Systems Co. will produce the system for the Chinese market. Opsys Technologies is also working with South Korean automotive components company SL Corp. to integrate its lidar systems into headlights and taillights.
With lidar a critical component of increasingly capable automated and autonomous driving systems, the market is expected to grow significantly.
Autobrains Technology explained its contrarian view on how artificial intelligence should work for autonomous driving at the EcoMotion conference. There’s too much emphasis on what is known as supervised learning, it said.
“Supervised learning means you need to really show every scenario the car will ever see,” said Joachim Langenwalter, the former Stellantis head of engineering who recently joined the Tel Aviv startup as senior vice president of autonomous driving. “Basically, it’s an encoder.”
Humans have to do the encoding, and it’s a massive, time-absorbing task to add new scenarios to the autonomous driver’s capabilities.
Autobrains Technology developed its AI to recognize patterns within the driving environment and require less human training. It segments information into clusters that it can use to identify patterns governing how it reacts to various driving scenarios.
Langenwalter said the company’s AI software is hardware-agnostic and can work with any combination of sensors and chip sets.