The field of autonomous driving has become a new arena for competition. How can traditional auto parts giants let it go?

The auto market has shifted from an incremental market to a stock market, and the era of high-speed growth is gone forever. Both automakers and parts companies are looking for new profit growth points. The field of autonomous driving has undoubtedly become a new arena for competition. As a traditional parts giant, Denso has already begun its layout in this field.

Denso established the ADAS Business Technology Development Department in 2016. The department is mainly responsible for determining the R&D roadmap for automated driving, as well as the development of technologies such as sensors and human-machine interface. In addition to its headquarters in Keria, Aichi Prefecture, Denso has established branches all over the world, such as Detroit, where North American automotive technology is concentrated, and the West Coast (Silicon Valley, San Diego), where innovative business models and IT companies gather, as well as Europe (Munich, Aachen, Linde, Gothenburg) and Singapore.

This article will provide a comprehensive analysis of Denso's autonomous driving development history, five key functions, partners, and investment layout in the field of autonomous driving.

Review of the development history of autonomous driving

1) Millimeter wave radar

Denso announced on August 10, 2017 that it has developed a 24-GHz submillimeter wave rear and side radar sensor to help enhance the functionality of vehicle safety systems. The 2018 Toyota Camry uses this radar sensor. Denso plans to place this radar sensor in the rear bumper of the vehicle to detect other vehicles that may enter the rear and side blind spots of the vehicle. In addition, the sensor can also detect other vehicles that may approach the rear of the vehicle when the vehicle is turning.

In 2015, "Toyota Safety Sense P", which installed an image sensor on Denso's millimeter-wave radar, was launched as an active safety kit for Toyota's mid-range and luxury models.

As early as 2003, Denso successfully developed an electronic scanning millimeter-wave radar sensor for automobiles, becoming the world's first manufacturer to use and commercialize digital beamforming technology.

2) LiDAR

As early as 1996, Denso developed a LiDAR for commercial vehicles. In 1997, Denso developed the world's first LiDAR, which can measure the round-trip time of light to one 2.5 billionth of the original, and can measure light beams in two dimensions, vertically and horizontally. By 2012, a miniaturized, low-cost linear LiDAR was commercialized and used in Daihatsu Motor's collision avoidance assistance system Smart Assist.

3) V2X airborne unit

Denso has successfully developed a high-speed LAN radio prototype since 1997 and obtained its first technical standard compliance certificate in Japan in 2000.

In 2006, Denso produced a 700 MHz band V2X test model (a verification test model based on Japan's frequency band allocation) to provide communication tests for Japanese companies, and it was successfully used in the determination of ARIB STD T109.

As mentioned above, DENSO produced prototypes and test models from the initial development stage and provided assistance in the determination of standard specifications. In 2015, DENSO successfully launched the world's first airborne equipment that meets its connectivity market.

Five key functions of autonomous driving

The first is sensing. By combining various sensors, such as millimeter-wave radar and visual sensors, Denso provides a perception system that goes beyond human vision, providing 360-degree perception capabilities, identifying surroundings even in harsh environments, and reducing blind spots and detection errors, as well as unnecessary operations caused by recognition failures.

The second is prediction. It can provide drivers with earlier and more accurate information about the road ahead than ever before. For traffic conditions beyond the driver's line of sight, the ADAS locator detects the location and driving route, and combines the information from V2X and DCM to understand the road conditions ahead. V2X/DCM is a communication measure that can correctly provide information beyond the visibility of the road ahead.

Among them, the ADAS locator can not only locate the position of the target vehicle and the driving lane, it can even detect the position and lane in the tunnel. The high-precision gyroscope and algorithm make detection possible.

The third is interactivity. The driver status monitor can detect the driver's condition and can immediately detect the driver's health and whether he or she is drowsy. The technology can also detect the direction of the face, even when part of the face is hidden by sunglasses and masks.

Fourth, connectivity. Denso increases driving comfort and fun by connecting vehicles to various external social information. The onboard communication unit connects with social backend systems, smartphones, etc., such as safe and comfortable automatic parking, multiple modes of smooth travel between vehicles and public transportation, and support for car software updates.

The fifth is protection. DENSO will add advanced cybersecurity measures to vehicles to protect vehicle systems from attacks. For example, the vehicle control system may be hacked and controlled by outsiders. DENSO has proposed a deep defense system to protect vehicles from threats from external control vehicles.

In addition, Denso's dual-drive EPS system, which consists of two independent electronic circuits and motor coils, usually generates power assistance from the dual-drive system. Even if one system fails, the steering function can be guaranteed.

如果有一个系统发生故障，EPS可以通过立即转向辅助其余的系统来继续保持平稳的驾驶，使转向功能的安全性得到显著提高。

Partner

In 2016, Denso and Toshiba jointly developed an artificial intelligence technology called Deep Neural Network-Intellectual Property (DNN-IP), which will be used in the image recognition system that has been independently developed by the two companies to help realize advanced driver assistance and autonomous driving technologies.

Deep neural network (DNN) is an algorithm that mimics the neural network of the human brain. It is expected to perform recognition processing as accurately as the human brain, or even more accurately than the human brain. In order to achieve autonomous driving, the car computer needs to be able to recognize different road traffic situations, including various obstacles, road markings, road conditions of driving roads, and potential dangerous situations.

In addition to cooperating with Toshiba, Denso also joined hands with Sony. On October 25, 2016, Denso announced that its automotive image sensors used products manufactured by Sony Semiconductor Solutions. The use of this sensor can improve the performance of the camera and identify pedestrians at night, which will help reduce nighttime traffic accidents.

Investment layout in the field of autonomous driving

1) Denso to acquire an additional 4.5% stake in Renesas Electronics for US$800 million

As automakers accelerate the application of autonomous driving and other technologies, in March 2018, Denso announced that it would acquire an additional 4.5% stake in Renesas Electronics for $800 million based on market prices. After the completion of this acquisition, Denso's stake in Renesas Electronics will increase to 5%, while INCJ's stake will fall to 45.6%.

2) Denso holds shares in Infineon

In November 2018, Denso acquired a stake in Infineon Technologies to accelerate its technology development and support the next generation of vehicle systems such as autonomous driving. After receiving the investment from Denso, Infineon Technologies' advanced semiconductor technology will be combined with Denso's vehicle technology and expertise to accelerate the development of new and emerging mobile application solutions.

3) Denso subsidiary NSITEXE announced investment in ThinCI

On September 6, 2018, local time, NSITEXE, a subsidiary of Denso, announced its investment in ThinCI. NSITEXE is a developer of key semiconductor components for autonomous driving, while ThinCI is an artificial intelligence (AI) hardware startup that develops computing platforms. ThinCI's programmable computing architecture can accelerate the development of deep learning, artificial intelligence, and other algorithms related to the automotive industry, and can increase the computing power and performance of the next best solution by 5 to 10 times.

4) Denso and NRI Secure set up a joint venture to provide testing and diagnostic services for connected cars

On September 27, 2018, local time, DENSO announced that it had agreed to establish a joint venture with Nomura Research Institute (NRI) Secure Technologies Co., Ltd. to detect and diagnose security risks of in-vehicle electronic products. The newly established cybersecurity company, named NDIAS Co., Ltd., will be jointly owned by DENSO and NRI Secure and will be officially established in December 2018. NDIAS will leverage DENSO's expertise in the development of in-vehicle quality and cybersecurity technologies.

5) Denso acquires Ease Simulation and acquires its on-board diagnostic technology

In November 2018, Denso acquired Ease Simulation, a company specializing in on-board diagnostic technology. Denso plans to merge Ease Simulation into Denso Products and Services Americas, which is committed to providing aftermarket products, services and technologies.

6) Denso’s stake in Infineon Technologies accelerates the development of autonomous driving technology

In December 2018, Denso announced that it would acquire a stake in Infineon Technologies AG to accelerate its technology development and support the next generation of vehicle systems such as autonomous driving. After receiving the investment from Denso, Infineon Technologies' advanced semiconductor technology will be combined with Denso's vehicle technology and expertise to accelerate the development of new and emerging mobile application solutions.

7) Denso joins hands with Toyota and SoftBank to invest $1 billion in Uber