Subscribe

ANSYS & Edge Case Research Looking to Transform Autonomous Vehicle AI

ANSYS & Edge Case Research Looking to Transform Autonomous Vehicle AI

Author: Eric Walz   

Software simulation company ANSYS (NASDAQ: ANSS) announced it is collaborating with Edge Case Research on the next generation of autonomous vehicles (AVs) to provide state-of-the-art perception capabilities. ANSYS is a global leader in simulation for a variety of industries, including automotive, aerospace and energy.

As part of a new agreement, Pittsburgh-based Edge Case Research, founded in 2013, will integrate its AV artificial intelligence (AI) perception stress testing and risk analysis system called Hologram, to support ANSYS' simulation solution to maximize the safety of AVs.

"We're excited to join the ANSYS AV ecosystem. We chose to partner with ANSYS because of their deep expertise in safety, which is critical to understanding how products like Hologram, VRXPERIENCE and SCADE can be used together in support of safety cases for autonomous products," said Mike Wagner, CEO at Edge Case Research. "ANSYS and Edge Case Research will deliver an unprecedented comprehensive capability for safeguarding the next generation of autonomous driving systems."

Edge Case Research will integrate Hologram with ANSYS' highly sophisticated AV open simulation solution. This end-to-end capability analyzes AV algorithms to detect edge cases and helps to validate perception algorithms in the most advanced autonomous driving systems.

 Although self-driving technology has become safer over the years, software engineers still must tackle the problem of "edge cases" which are extremely rare events that might be encountered on the road. Although these events are rare, they still must be addressed to improve the safety of self-driving vehicles. 

Autonomous vehicles being developed today are "trained" to recognize and respond to changes in road conditions by running though thousands of scenarios, and more importantly developing processes to respond to rare edge cases. 

ANSYS' simulation software allows parameters to be tweaked to aid in testing.

AI perception systems being developed for self-driving vehicles may fail to detect hazardous edge cases because deep-learning algorithms have not been trained for every possible road situation it might encounter in the real world.  ANSYS provides developers with the tools to automatically identify these challenging edge cases, which the company says are far more scalable than manual data labeling. 

Edge Case Research developed Hologram to help in the designing and training of machine learning algorithms by complementing traditional computer simulation and road testing of perception systems. It intelligently tests perception software against adversarial examples, which it generates from a customer's own sensor data. 

Hologram identifies risks that are difficult to find with other types of testing and analysis and shows how perception software reacts in realistic "what if" scenarios.

After perception data is collected by a self-driving vehicle, Hologram helps find the edge cases where perception software exhibits odd, potentially unsafe behavior. This gives developers a chance to address these scenarios before they affect safety and performance, thereby improving autonomous vehicle perception systems. 

"Edge Case delivers a powerful data testing and analytics platform that unlocks the value of petabytes of AV's recorded road data to find edge cases, significantly accelerating the development of safer, AI-driven perception software. Underlying capabilities have been incorporated into our recently announced collaboration with BMW," said Eric Bantegnie, vice president and general manager at ANSYS. "Together we will usher in a new era of AI and shape the future of safe autonomous driving."

In June, Edge Case Research announced a $7 million funding round led by Chris Urmson, the former CTO of Google's self-driving car program and the co-founder and CEO of autonomous driving startup Aurora. Urmson is also a member of the Edge Case Research board of directors.

"Self-driving technology will have a profound impact on the safety of our roads and the quality of life of communities in our cities,"  Urmson said in June. "Developing and delivering self-driving technology safety is paramount, and the Edge Case team has the commitment and deep expertise for building important tools for developing this technology.  

Eric Walz
Eric Walz
Originally from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry in Silicon Valley. Eric has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology. Outside of work, Eric likes to travel to new places, play guitar, and explore the outdoors.
Recommended
Prev:Japan & France to Explore Cooperation on Automotive Technology, Including EVs Next:BMW i Ventures Announces Investment in Cartica AI
Comment
    view more