Cisco Blogs
Share

AImotive: Can Self-Driving Cars be as Safe as Airplanes?


March 22, 2018 - 3 Comments

Riding in a plane is significantly safer per mile than traveling by road. For the majority of people, driving or being a passenger in a motor vehicle is the most dangerous action they will perform in a day. Fatalities caused by aviation have steadily declined since the 1960s, but road safety has remained almost steady and significantly worse than other modes of transportation.

Can we make motor vehicle safety as safe as riding an airplane? AImotive, the full-stack autonomous vehicle technology company based in Budapest, Hungary, says YES. The aviation industry has used simulation technology for decades to train their pilots, AImotive is now developing simulation test environments to do the same thing for self-driving cars. “We want to reach similar levels of safety on the roads as have been reached in the skies,” said Laszlo Kishonti, founder and CEO of AImotive.

The auto industry is heading full bore toward self-driving vehicles, with a goal of reaching “level four” — truly autonomous vehicles — by 2020. AImotive expects to be a major player providing solutions for bringing the dream of autonomous vehicles to reality. The company’s custom software allows some of the necessary vehicle testing to be done through simulations, reducing the amount of road testing needed to cover a variety of weather and road conditions.

As an example, simulation testing can be used to determine whether the autonomous driving software would be able to detect a person lying in the middle of the road in snowy conditions. With tight regulations governing where autonomous cars can be tested, it’s not always possible to test all specific real world scenarios. Simulations provide a way to check all corner cases and a multitude of scenarios at a fraction of the cost.

AImotive has announced the closing of its $38 million USD Series C funding round, led by B Capital Group and Prime Ventures, with participation from Cisco Investments, Samsung Catalyst Fund, and Robert Bosch Venture Capital, Inventure, Draper Associates and Day One Capital. The company will use the new capital to continue developing its autonomous vehicle driving software stack. AImotive uses a camera-first, AI based computer vision technology to help self-driving cars recognize and avoid obstacles. Its software integrates visual data from low-cost camera sensors, and also uses sensor fusion to incorporate input from other sensors to provide greater safety in varied weather and environmental conditions.

AImotive’s solution is designed for automotive OEMs, Tier 1 suppliers and other mobility service providers, and the company is working with some of the established automotive players including Groupe PSA, SAIC, and Volvo.

We are excited that AImotive represents Cisco Investments’ first pure AI-focused investment in Europe. Since we realize that Cisco’s next generation of applications, products and services will be enabled by AI, we look forward to building on this investment theme in the future, with a particular emphasis on unlocking the power of data, computer vision, robotics, intuitive networks. If your early-stage company is working on these types of technology, feel free to reach out to Cisco Investments to discuss the potential for a partnership.



In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.

3 Comments

  1. Personally, I do not think the tech and AI are well developed for self-driving cars. The best example of the same is uber self-driving car killing a person. More needs to be worked on AI so that it can understand and identify humans as humans. http://tecknowlogic.com/

  2. A very interesting topic. If we have a fully-autonomous car network, we would expect all the connected cars will be continuously communicating to each other and performing negotiation. With aviation background, I had always remember a quote within the industry 'Whose is flying the air plane'. This is very much apply to all cases, and especially to automated pilot system in the modern aircraft. Often pilots are forgetting that they should be in control of the aircraft, and not the auto-pilot computer system.

  3. I think fully automatic IoT enabled cars/vehicles would not be as much more safer. However, there is still a lot of development needed to be made. If you are curious about driver safety & performance do check out this link: https://aware360.com/driver-performance/