The agreement outlines the intent to research and develop solutions to emerging challenges related to processing performance, power consumption and design complexity.

IBM (NYSE: IBM) and Honda Motor Co., Ltd. (Honda) announced the signing of a Memorandum of Understanding (MOU) outlining their intent to collaborate on long-term research and development of cutting-edge computing technologies[1] needed to overcome challenges related to processing power, power consumption and design complexity to create software-defined vehicles (SDVs) of the future.

The application of artificial intelligence (AI) technologies is expected to accelerate significantly in 2030 and beyond, creating new opportunities for the development of software-defined vehicles. Honda and IBM predict that SDVs will dramatically increase the complexity in design, processing performance, and corresponding power consumption of semiconductors compared to conventional mobility products. To solve anticipated challenges and achieve highly competitive software-defined vehicles, it is critical to develop independent capabilities in research and development of next-generation computing technologies. Based on this understanding, the two companies began to consider joint long-term R&D opportunities.

In particular, the MOU outlines areas of possible joint research into specialized semiconductor technologies, such as brain-inspired computing.[2] and chip technologies, with the aim of drastically improving processing performance while decreasing energy consumption. Joint optimization of hardware and software is important to ensure high performance and rapid market launch. To achieve these benefits and manage complexity in the design of future SDVs, the two companies also plan to explore open and flexible software solutions.

Through this collaboration, both companies will strive to create software-defined vehicles that deliver world-class computational performance and energy savings.

[1] Computing technology is developing with the aim of achieving high processing performance and low power consumption in the 2030s and beyond.

[2] Computational architecture and algorithms that mimic the structure and function of the brain while optimizing silicon.

quick access