BMW to introduce self-driving vehicle by 2021
BMW, Intel, and Mobileye have added Delphi to their partnership to create a platform that the industry can use to develop self-driving vehicles. For BMW, it is a step towards having an autonomous vehicle by 2021.
It is not easy, and perhaps not safe, for every automaker to try to develop autonomous vehicles from scratch. Most are working on the technology and have an array of sensors, cameras, radar, and millions of lines of code in the computing systems on board vehicles already today. But to produce level 3-5 autonomous vehicles—meaning the vehicles can at the least make their own decisions to speed up and change lanes and at the most can be designed for full automation with no human, pedals, or steering wheel necessary—involves complicated technology and systems.
In the intertwined global industry, Delphi already is working on a project with Mobileye which was recently acquired by Intel. And Delphi has worked with BMW for years. This particular collaboration brings the four players together in a partnership to develop a platform that will be open to the industry as a whole and will help BMW meet its goal of having its iNext autonomous vehicle on the road by 2021.
The partnership doesn’t deliver an autonomous car, but the technical platform to build one, said Richard Rau, who is in charge of the sensors, control units, and software units of BMW’s autonomous driving department.
Delphi is the integrator, and the platform will be flexible enough that it can be modified and customized for any customer. Delphi’s deal is not exclusive. BMW can also tap other system integrators.
Rau said he is in deep discussions with other automakers and new partners will be revealed in a few weeks. In cases where automakers are advanced in their own work in this area, the platform can be adapted to accommodate where they are in their own development and help accelerate the process. General Motors, Mercedes-Benz, Ford, Volvo, Audi, and many others have spent many years already working on autonomous vehicles. The more the merrier as economies of scale will help bring down the high costs of this technology and the expectation is there will be consolidation in this field in the future. The industry also must ensure these vehicles are safe, and collaboration and standards are one way to do so.
Delphi offers the group hardware such as sensors and electrical architecture as well as software, integration of all of the above, and a data component which is essential as self-driving vehicles rely on a lightning-fast stream of data that allows a vehicle to read its surroundings and make driving decisions. The data must be validated and quickly pass to and from the storage cloud.
In July, BMW, Intel, and Mobileye announced they were working together to bring self-driving vehicle technology to production by 2021. They have since developed a scalable architecture that can be used by BMW as well as other automakers and suppliers working on autonomous technology. Delphi has already provided a prototype computing platform to BMW that integrates sensors and other key components into BMW’s own vehicle architectures.
Delphi Chief Technology Officer Glen De Vos said the next few months will be used to sort out who does what and how to create a computing platform that any automaker can use to get autonomous technology into their own vehicles.
Kathy Winter, vice president of Autonomous Driving for Intel, said the group offers other automakers the opportunity to purchase a computing platform and data center for artificial intelligence and advanced learning. The technology is key to the intelligence needed if the vehicle is in charge instead of a human.
Delphi has been focused on its ability to handle big data, making acquisitions of startups in this area in anticipation of a huge growth in demand for high speed data transmission to and from vehicles. The supplier also announced earlier this month that it was dividing into two entities, spinning off the powertrain division devoted to combustion engines from the part devoted to electrification and automation including autonomous vehicles.