Toyota Invests in Virtual AV Software Dev Parallel Domain

Toyota AI Ventures, the venture capital subsidiary of the Toyota Research Institute (TRI), is investing in Parallel Domain, a Silicon Valley-based software developer specializing in 3D environment generation for autonomous vehicle simulation.

The company joined the $2.65 million seed round led by Costanoa Ventures and Ubiquity Ventures, with participation from RRE Ventures, and Bessemer Venture Partners, which was announced earlier this month.

Founded in 2017, Parallel Domain helps developers to generate a wide variety of highly detailed simulation environments and dynamic scenarios based on both real and fictional locations.

“Generating 3D environments and scenarios has become a primary bottleneck. Scaling this generation is a specifically difficult problem that we’re uniquely equipped to solve,” Parallel Domain founder Kevin McNamera wrote in a December 6 blog post.

Among the software capabilities the company touts is the ability to build a world that changes over time. For instance, adding bike lanes, scattering garbage, repaving roads with new asphalt or advancing time so trees grow or roadways weather and crack.

Virtual testing is seen by software developers, research institutes and automakers alike as a way to further develop artificial intelligence (AI) systems embedded in self-driving software.

The theory goes, the more scenarios — and virtual miles — the software has to draw from, the more prepared the self-driving vehicle will be for a variety of road conditions, weather events and other variables.

Open source software could also help developers build a data pool from a wide variety of sources. In June TRI invested $100,000 in CARLA, an open source simulation project to take advantage of the collaborative development potential of an open source community.

Parallel Domain also faces stiff competition from tech giants such as Nvidia, which recently signed an agreement with British drive simulation software specialist rFpro.

The company will provide the virtual environment used to develop AVs in the Nvidia Drive Sim and Nvidia Drive Constellation products.

Earlier this week, simulation company Ansible Motion unveiled its latest laboratory environment, which it claims will help carmakers assess human interactions with advanced driver assistance systems (ADAS) and AI platforms.

In addition to offering a wide variety of weather and other environmental scenarios, the company claims its Delta Driver-in-the-Loop (DIL) simulator can also handle behavioral differences depending on different countries’ cultural backgrounds.

Meanwhile, Hyundai Mobis announced that it plans to use 3D video game development software to develop a simulated driving environment for AVs, and Microsoft’s Project Road Runner team has been using photo-realistic simulation and deep learning to train autonomous-driving algorithms.

In September, German auto-giant BMW announced construction of a facility for the simulation of real-world driving situations in the north of its headquarters-city, Munich.

One of the central features of the facility is the high-fidelity simulator, in which longitudinal, transverse and rotational movements of a vehicle can be represented simultaneously and, hopefully, more realistically.

Among the other major players investing in virtual AV testing include Autonomous Intelligent Driving (AID), a subsidiary of German auto giant Audi, which is partnering with Cognata.

The Israeli tech firm leverages artificial intelligence, deep learning and computer vision within a realistic simulation environment to judge and validate AVs prior to physical roadway tests.

— Nathan Eddy is a filmmaker and freelance journalist based in Berlin. Follow him on Twitter.


Leave a comment

Your email address will not be published. Required fields are marked *