Integrating Linux into automotive systems for the long haul
The Linux Foundation is revving up deployment of its widely used operating system by launching the Automotive Grade Linux Work Group to advance automotive system development through collaboration between the Linux and open-source communities and the automotive industry. Rudolf explains how the project will fuel in-vehicle technology innovation by developing a reference platform that meets the automotive industry's needs for connectivity and long-term viability.
ECD: What is the Automotive Grade Linux (AGL) Work Group, and what are its objectives?
STREIF: The AGL Project was created in response to demand from Linux Foundation members in the automotive industry – members like Toyota, Jaguar Land Rover, HARMAN, and others. It is a collaboration platform that fosters the adoption of Linux and open-source software for automotive applications. The project’s scope includes but is not limited to the development and support of a reference Linux Operating System (OS) software stack, selection and verification of reference hardware for that OS stack, development and build processes and tools for Linux systems, middleware and applications for automotive systems, education and tools for open-source compliance, and supply chain management. The project provides a full Linux OS stack for in-vehicle applications and related processes, tools, and documentation to implement a reference design meeting automotive requirements.
ECD: Why was the Tizen platform selected for vehicle-based Linux applications?
STREIF: Tizen In-Vehicle Infotainment (IVI) is a good starting point for AGL. It offers technology on top of which the group can build and innovate for the automotive industry. The platform also provides the necessary operating base infrastructure and incorporates essential software packages for embedded connected devices.
AGL is leveraging existing efforts and adding value to them. AGL work will be done in upstream open-source projects, and Tizen IVI is one of many possible ways to integrate these projects into a fully functional OS stack. Carmakers and their suppliers are free to choose the integration form that best fits their technical and business requirements.
ECD: With the typical automobile lifespan measured in decades, how can embedded in-vehicle computer systems keep up with the rapid changes in technology?
STREIF: The short answer is: with Linux and collaborative development. No other OS or development model can support the long-term needs of carmakers. The massive community and multiple vendors supporting this ecosystem ensure long-term support. This is why carmakers are flocking to Linux and open-source software to support their technology requirements. The goal of AGL is to keep pace with quickly evolving technologies by developing and integrating them. The AGL stack provides the most cutting-edge technical features for engineers to experience and evaluate up-to-date technologies. To accommodate 10-plus years of the typical automotive product life cycle, AGL will cooperate with The Linux Foundation’s Long-Term Support Initiative for the Linux kernel ( ). The AGL software stack will also include sophisticated software for over-the-air software and firmware updates as well as remote device and software management. This will allow carmakers to update and maintain in-vehicle software remotely, providing vehicle owners with a great user experience over the vehicle’s lifespan.
ECD: Since cloud connectivity is at the center of in-vehicle designs, what security technologies are available to avert unwanted intrusions?
STREIF: Security is a major priority of AGL. There are several areas that need to be addressed: platform integrity, user privacy and user data protection, and digital rights management and conditional access to third-party content. Security will be designed into AGL hardware and software platforms, which will utilize cryptographic hardware and software to secure the platform, data, and content. For instance, software, including but not limited to third-party applications, will need to be signed for authenticity and encrypted before delivery. User data such as location information will be sent encrypted over secure connections to cloud services to prevent attacks. User data stored on the device such as phone directories will be secured so that only the authorized user can access it.
ECD: What is your vision for the future of in-vehicle infotainment technology, and what challenges stand in the way?
STREIF: Our vision is to accelerate in-vehicle system technology development by creating a collaboration platform for the industry to leverage Linux and open source for product development. The possibilities are endless, and many new services and business models are expected to arise. For vehicle drivers, we envision services such as updated road information directly retrieved from sensors that instrument the roads, intelligent traffic management systems that direct drivers around traffic jams using different routes according to actual road load, vehicle-to-vehicle communication to alert of potential hazards, and energy management for electric cars. For passengers, we see live content from the Internet, plus video on demand for movies, live television, office applications, and more. The connected car will be an integrated part of the consumer lifestyle just like the smartphone is today. Users will have the ability to seamlessly access cloud content as well as continue to watch television shows moving from the home to the car.
Ultimately, the connected car will require an infrastructure to support it, and without it will remain nothing but a vision. However, building and maintaining an infrastructure is costly. Public and private entities will have to invest large sums, and eventually consumers will have to foot the bill in one way or another. Only open systems and open interfaces will be able to create the necessary economies of scale and network effects that will make the investment worthwhile.
Open does not mean that access and content are free; the truth is far from it, but open does mean interoperability. No matter where the road takes us, we’ll want ubiquitous access to our favorite content and services.