Home News OTSL strengthens its 3D real-time sensor simulator framework ecosystem for autonomous driving

OTSL strengthens its 3D real-time sensor simulator framework ecosystem for autonomous driving

US: OTSL has established partnerships with three companies providing advanced technologies related to electronic systems and autonomous driving: Germany’s COSEDA Technologies GmbH; U.S.-based VectorZero; and Finland’s AILiveSim. These partnerships will enable OTSL to strengthen the ecosystem for its 3D real-time sensor simulator framework COSMOsim (R), which is aimed at autonomous driving. COSMOsim (R) is the only platform in the world that can simultaneously run five types of simulators — corresponding to millimeter wave radar, LIDAR, cameras, infrared light, and ultrasonic wave sensors — in real time on a single screen.

As autonomous driving continues to move toward practical use around the world, there is increased demand for powerful simulation technologies that can virtually recreate diverse driving conditions to enable safe and accurate verifications.

“Achieving completely autonomous driving requires moving beyond vehicle-mounted sensors and automotive systems such as Electronic Control Units (ECUs) and in-vehicle networks, and creating an environment in which the countless factors involved in automotive societies — roads, the surrounding environment, human engineering, and more — can be simulated with a high level of accuracy,” said Shoji Hatano, CEO, OTSL. “By partnering with global companies that offer the most cutting-edge technologies to construct the ecosystem, OTSL is aiming to create an advanced autonomous driving simulation environment as soon as possible.”

Partnering with Germany’s COSEDA Technologies GmbH enables the development of autonomous driving systems that link COSEDA’s electronic system design tool COSIDE (R) with COSMOsim (R). An autonomous driving system simulation environment can be built on COSIDE (R) which uses IEEE Std 1666.1-2016 SystemC AMS language, the global market leader in design tools for Software in the Loop Simulation (SILS) environments, where all simulations are based on software, ranging from semiconductors to vehicle systems.

There is no need to prepare automobile hardware such as sensors and ECUs. Everything can be simulated through software, which significantly reduces development time and costs. COSEDA Technologies’s system-level simulation technology furthermore enables the verification of sensors and ECUs in the initial stages of design, functional safety factors and compliance with standards in near future, greatly enhancing the efficiency of development.

COSMOsim (R) is now compliant with ASAM OpenDRIVE, a file format for describing the form of road networks necessary for driving simulations. ASAM OpenDRIVE is managed by the international automation and measuring system standardization group ASAM e.V (the Association for Standardization of Automation and Measuring Systems).

Through the partnership with U.S.-based VectorZero, COSMOsim (R) can now read and write ASAM OpenDRIVE files using RoadRunner, VectorZero’s road modeling and traffic simulation tool. This enables COSMOsim (R) to recreate more realistic road conditions — intersections, bridges, and even road signs. Complying with international road network standards means that OTSL can offer a highly accurate simulation environment to companies in the global autonomous driving market.

Finally, by partnering with Finland’s AILiveSim, COSMOsim (R) can now link with the next-generation AILiveSim autonomous driving simulation platform, which employs deep learning technology. In addition to autonomous driving simulations using automobiles, structures, and the like, AILiveSim has even moved into the field of maritime travel.

With this partnership, COSMOsim (R)’s diverse sensor simulation techniques can be applied to maritime traffic as well, enabling OTSL to expand the business field of its autonomous driving simulator products. AILiveSim and COSMOsim (R) use the same game engine, Unreal Engine 4, and are therefore highly compatible. This makes it easy to effectively use AILiveSim’s main characteristics — AI and deep learning — in COSMOsim (R) and achieve a level of performance in autonomous driving simulations not yet seen.