Qualcomm, the dominant provider of cellular modem chips, is now trying to conquer the auto-driving market. Qualcomm at Monday’s Consumer Electronics Show announced its Snapdragon Ride platform, which uses chips from Qualcomm’s mobile products to help drive the car.
It may seem strange that a smartphone chip company is involved in auto-driving, but in fact, many technologies are the same in both products. Qualcomm has long been using its mobile chips to break into the market, not only among smartphones, but any technology that uses them. Smartphone manufacturers are buying Qualcomm Snapdragon chips, which contain an ARM-based processor, a modem chip, and various auxiliary chips.
In recent years, Qualcomm and other mobile chip manufacturers have begun to include in their products system-on-a-chip (SoC) more and more powerful graphics processors and special artificial intelligence chips – this is the basis for developing automotive software based on machine learning. Thus, the launch of the new SoC, which includes (as stated in the Qualcomm press release) high-performance multi-core processors, energy-efficient artificial intelligence and computer vision engines, industry leading graphics accelerators, is not such a big leap for Qualcomm.
Insiders usually divide self-driving technology into two levels. Some companies are working on the creation of fully self-propelled systems that can work without human intervention – although at present they are mainly prototypes, the safety of which is controlled by people. In addition, there are modern driver assistance systems (ADAS). They are designed for driving in limited conditions – most often on highways – and only under the supervision of a person. Most major automakers already ship ADAS systems, and over time they become more sophisticated.
Qualcomm says the Snapdragon Ride platform includes much more than just hardware. The platform will come with cross-platform software, operating systems and drivers for the chips. In addition, Qualcomm will offer software for localization, perception and prediction of car behavior, and these are three key elements in any autonomous driving system.
Qualcomm says it initially targets the ADAS market. The company claims that its new system can provide 30 trillion operations per second (TOPS), attracting less power for this than competing chips. Qualcomm says it also plans to release more powerful systems that can carry out 130 trillion operations per second, suitable for applications for fully autonomous driving.
Low power consumption is expected to be one of Qualcomm’s selling points. Energy consumption is obviously an important feature of mobile devices, as customers expect devices to work all day without recharging. The power budget is not so limited on board a running car, but energy consumption is still of great importance. If Qualcomm can provide sufficient computing power to deploy some of the energy-efficient technologies developed in the mobile world, the company will be able to offer an attractive platform for automakers.
However, Qualcomm rivals – primarily Nvidia and those owned by Intel Mobileye – are not standing still. Nvidia has long provided enormous amounts of processing power for its conventional graphics cards, which is well positioned to serve the computationally-hungry autonomous driving market.
Qualcomm’s announcement includes an inaccurate statement that they will work with GM on ADAS systems, but so far there is no mention that GM uses chips in any particular vehicles. Qualcomm says they are keen to introduce Snapdragon Ride chips in car manufacturing in 2023.
Advantages of Snapdragon Ride
- Integrated transport security suite with reliable OS and hypervisors
- Security systems from various companies including Adaptive AUTOSAR
- Optimized libraries of basic functions for computer vision systems, signal processing, standard arithmetic libraries
- AI tools on heterogeneous computing systems
- Full-featured software stack of functions for driving on the track, including perception and planning
- Qualcomm Vision Enhanced Precise Positioning (VEPP) Cost-Effective Localization Solution
- Software and hardware for the testing process
- Data management tools for intelligent data collection and automated annotation
We are perhaps the most powerful competence center in Russia for the development of automotive electronics in Russia. Now we are actively growing and we have opened many vacancies (about 30, including in the regions), such as a software engineer, design engineer, lead development engineer (DSP programmer), etc.
We have many interesting challenges from automakers and concerns driving the industry. If you want to grow as a specialist and learn from the best, we will be glad to see you in our team. We are also ready to share expertise, the most important thing that happens in automotive. Ask us any questions, we will answer, we will discuss.
Read more useful articles:
- Cameras or lasers
- Autonomous cars on open source
- McKinsey: Rethinking Software and Electronics Architecture in Automotive
- Another OS war is already under the hood of cars
- Program code in the car
- In a modern car, there are more lines of code than …