From ADAS to unmanned key technology what are the _adas_ auxiliary driving _ automatic driving

Mr. Li announced that Baidu would launch an unmanned vehicle this year, and the chief executive of Tesla said shortly ago that "driverless cars are not a thing", meaning that the era of unmanned driving is coming. But in fact, unmanned driving is still in the initial stage of the dream, Google's unmanned vehicles have long been on the road test, so far still not commercially, that is evident. However, prior to realizing the dream of unmanned driving, the Advanced Driver Assistance System (ADAS) can actually give the driver a real benefit. Adas can assist and complement the driver in the complex vehicle control process, and will eventually achieve unmanned driving in the future.



Advanced Driver Assistance System (Advanced Driver Assistantsystem), referred to as Adas, is the use of a variety of sensors installed on the vehicle, in the first time to collect the vehicle inside and outside the environment data, static, dynamic object identification, detection and tracking technology processing, This allows drivers to detect potential hazards at the fastest time to attract attention and improve security of the active security technology. ADAS uses sensors that include cameras, radars, lasers, and ultrasound to detect light, heat, pressure, or other variables used to monitor the state of a car, usually in front and rear bumpers of the vehicle, in-side mirrors, inside the driver's shaft, or on the windshield. The early Adas technology is mainly passive alarm, when the vehicle detects the potential danger, it will alert motorists to the unusual vehicle or road conditions. Active intervention is also common for the latest Adas technologies.

Key technology and application of ADAS system

Adas's two key technologies are processors and sensors, and while ADAS applications are becoming more complex, adas applications are being popularized from luxury high-end cars to middle and low cars as the cost of device performance increases. For example, adaptive cruise control, blind spot monitoring, lane departure warnings, night vision, lane maintenance assistance, and collision warning systems, active ADAS systems with automatic steering and braking intervention are also starting to be used in the broader marketplace.



System: Lane departure alarm sensor: Camera

When a vehicle leaves its driveway or approaches the edge of the road, the system emits sound alarms or action alarms (via a slight vibration steering wheel or seat). When the vehicle speeds exceed a certain threshold (for example, greater than 55 miles), the vehicle does not turn on the turn signal, and these systems will begin to play a role. When a vehicle travels, its position relative to the lane mark indicates that the vehicle is likely to deviate from the driveway, and the camera system is needed to observe the lane sign. Although these application requirements are similar for all vehicle manufacturers, each vendor uses a different approach, using a front-view camera, a rear-view camera, or a dual/stereo camera. For this reason, it is difficult to adopt a hardware architecture to meet the various types of camera requirements. A flexible hardware architecture is needed to provide different implementation options.



System: Adaptive cruise Control sensor: Radar

In the past decade, luxury cars have adopted the ACC (Adaptive cruise Control) technology, which is now being used in a wider market. Unlike the traditional cruise control technology designed to keep vehicles at constant speed, ACC technology makes the speed and traffic conditions compatible and, if too close to the front, slows down and accelerates to the upper limit when the road is allowed. These systems are implemented by using a radar that is mounted in front of the vehicle. However, because the radar system can not recognize the size and shape of a target, and its field of view is also relatively narrow, so the application should be combined with the camera. The difficulty is that the cameras and radar sensors currently in use do not have a standard configuration. Therefore, a flexible hardware platform is required.



System: Traffic sign identification sensor: Camera

As its name shows, the traffic sign recognition (TSR) feature uses a forward camera combined with pattern recognition software to identify common traffic signs (speed limit, stop, turn around, etc.). This function will remind the driver to pay attention to the traffic signs in front of them so that the driver can follow them. TSR reduces the driver's failure to comply with traffic regulations such as parking signs, and avoids illegal left-turn or unintentional other traffic violations, thereby improving security. These systems need a flexible software platform to enhance the detection algorithm, according to different areas of traffic signs to adjust.

System: Night Vision Sensor: IR or thermal imaging camera

The Night Vision (NV) system helps drivers identify objects under very dark conditions. These objects generally exceed the field of sight of the headlights, so the NV system gives early warning to the vehicle on the road ahead to help the driver avoid the collision. The NV system uses a variety of camera sensors and displays, specific to the manufacturer, but generally belongs to two basic types: active and passive. The active system, also known as the near-IR system, combines a live coupled device (CCD) camera with an IR light source to render a black-and-white image on the monitor. The resolution of these systems is very high and the image quality is very good. Its typical visual range is 150 meters. These systems can see all the objects within the camera's field of view (including objects without heat radiation), but in the rain and snow environment, the efficiency is greatly reduced. Instead of using an external light source, the passive system relies on a thermal camera to collect images using natural thermal radiation from the object. These systems will not be affected by the headlights on the opposite side, and will not be affected by bad weather conditions, the detection range reached 300 meters to 1000 meters. The disadvantage of these systems is that the images are granular and functionally limited to warmer weather conditions. Moreover, passive systems can only detect objects that have thermal radiation. Passive system combined with video analysis technology, can clearly show the vehicle in front of the road objects, such as pedestrians. In the NV system, there are a variety of architecture choices, each of which has its advantages and disadvantages. To improve competitiveness, automotive manufacturers should support a wide range of camera sensors to implement these sensors on a versatile, flexible hardware platform.



System: Adaptive far-light control sensor: Camera

Adaptive far-light control (AHBC) is an intelligent headlight control system that uses a camera to detect traffic conditions (the opposite vehicle and the same traffic condition) and, depending on these conditions, lightens or dims the far light. The AHBC system supports the driver to use the remote light as far as possible at the maximum illuminating distance without manually dimming the headlights when other vehicles appear, without distracting the driver's attention, thus improving the vehicle's safety. In some systems, you can even control the headlights, dimming a large lamp, while the other headlamps are lit normally. AHBC and LDW and TSR front camera systems are complementary. These systems do not require high-resolution cameras, and a vehicle that is already using a front-view camera in ADAS applications can have a very high price/performance ratio.

System: Pedestrian/obstacle/vehicle detection (PD) Sensors: Camera, Radar, IR

Pedestrian (as well as obstacle and vehicle) detection (PD) systems rely entirely on camera sensors to deeply perceive the surroundings, such as using a camera, or using a stereo camera in a more complex system. The difference between "category variables" (dress, lighting, size, and distance) is large, complex and changing, and the sensor is positioned on the mobile platform (vehicle), which makes it difficult to determine the visual characteristics of the pedestrian in the mobile, so the PD system can be enhanced by using an IR sensor. Radar can also enhance the vehicle detection system, which provides a good distance measurement function, in bad weather conditions, performance outstanding, can measure the speed of vehicles. This complex system requires the use of data from multiple sensors at the same time.

System: Driver Sleepy Alarm sensor: in-car IR camera

The sleepy alarm system monitors the driver's face, measuring its head position, eyes (open/closed), and other similar alarm instructions. If the driver is determined to have signs of sleep or appears unconscious, the system will issue an alarm. Some systems also monitor heart rate and respiration. A feature that is conceived but not yet implemented involves driving the vehicle close to the curb and eventually stopping.