Israeli start-up mimics nature to enable autonomous vehicles to see

In the race to develop fully autonomous vehicles, being able to accurately “see” the road ahead and assess potential hazards is critical for safe navigation.

While most leading industry actors have relied on and heavily invested in laser-based LiDAR (light detection and ranging) three-dimensional sensors for self-driving navigation, Tesla CEO Elon Musk has been the primary – and vocal – proponent of navigation based on using inexpensive cameras and radar.

While developers continue to argue among themselves regarding the pros and cons of the two systems, Rehovot-based robotic vision start-up Lirhot Systems says it has developed a third method of navigation: a camera-like sensor inspired by insect navigation.

“In nature, you have bugs and insects that navigate in a specific way, and we’re copying that to enable autonomous vehicles to see,” Lirhot CEO Shlomi Voro, an applied physicist with dozens of patents in the field of quantum physics, told The Jerusalem Post.

“We were inspired by the heads of bees, their artificial intelligence-like neural network, size, accuracy of navigation, and how they see the world through their five eyes – two for vision and three for navigation.”

Established in 2017 after spinning off from design and prototype development company Vorotec, Lirhot has raised $2 million in funding to date from Rafael Development Corporation, the commercial arm of Rafael Advanced Defense Systems.

The company’s in-house developed 5D artificial intelligence-based vision imager, Voro said, can provide the accurate location of an autonomous car, ship, aircraft or any other robotic application in motion anywhere on the planet.

Navigating only by using atmospheric light and rapidly identifying true north, the system does not rely on outside communication, including GPS or any other radio communication means.

“Everyone knows what happens when you start off navigating with Waze, when it takes time to understand your location and which direction you are facing,” said Voro.

“When you are the driver, you just take the next turn and get back on track. When it’s an autonomous vehicle, you don’t have that privilege. You need to know where you are, accurately, at all times.”

The bio-inspired technology also enables precise classification of materials and substances on the road surface – such as water, oil or ice – that could pose a threat to the driver and passengers if not accurately identified.

Since the turn of the year, Lirhot has been conducting a proof-of-concept (POC) with a “large customer,” and is seeking additional partners, primarily in the field of cyber navigation – free from the threat of GPS spoofing.

“We essentially took the insect’s head and put it on one piece of hardware, which you just connect to your car or vehicle,” Voro said.

“This enables us, first, to be able to drive autonomously, and second, to prove that a camera-like system is far more advanced than many LiDAR systems on the market.”