Home > Hot News

Hot News

Contact us

Marketing Department
Contact£º David
Tel£º +86-631-3658331
Email£º sc05@e-dongxing.com
Address£º NO.152,Shenyang Road, Hi-tech Industry Dev, Zone, WeiHai ,Shandong ,China

Company News

Robotic cars and sensing technology bring us closer to autonomous driving - From TDK
Time£º2016-7-20 8:10:51¡¡¡¡¡¡Author£ºMR.wang¡¡¡¡¡¡Clicks:

People, things, and information linked by networks¡ªthis is the age of the cloud. Along with the advancement of ITS (Intelligent Transport Systems) and telematics, automobiles and roads are increasingly networked as well, and the development of autonomous driving systems is progressing at a rapid pitch in various locations around the world. One of the key technologies towards the realization of safe and pleasant autonomous driving is sensing technology that allows precise acquisition of vehicle position and peripheral information. Various kinds of highly sensitive and efficient sensors taking the place of the five human senses, along with information processing technology and artificial intelligence (AI) are indispensable for the emergence of the robotic car.

The robotic car of the near future¡ªa mobile office and mobile living room

The convenience of not having to drive oneself is not the only advantage of a robotic car. The number of people killed in traffic accidents worldwide exceeds one million. About half of these are pedestrians or persons riding a bicycle or motorbike, and a large percentage of accidents are caused by human error. With an autonomous driving system that constantly monitors and scans the situation around the vehicle, failure by the driver to notice something, wrong assumptions, overconfidence in one's driving abilities and other types of human errors can be avoided, which should result in a drastic reduction in the number of traffic accidents.

Advances made in the areas of sensing technology and information processing are of great significance with a view towards making robotic cars a reality. Some automobiles already incorporate ACC (Adaptive Cruise Control) systems that can maintain a proper distance from a preceding car, automatic braking systems for collision avoidance, lane keeping assist systems, and other assistive features that partially approach the realm of autonomous driving.

Robotic cars have currently reached the level where experiments are being carried out on public roads, with practical realization being targeted for the second half of 2020s. The cars of the near future are envisioned to be more like moving offices and moving living rooms. Once autonomous driving is also realized for trucks, the patterns of motorization and goods distribution are bound to undergo a major change.

Laser radar to serve as highly capable "eyes" of the robotic car

An automobile comprises a large number of many kinds of sensors, which can be categorized for example as propulsion, body, safety, and info-communication related. Sensors that monitor the vicinity of the vehicle for accident prevention use various technologies such as millimeter wave radar, infrared radar, optical cameras etc. Millimeter wave radar using radio technology can identify objects at fairly large distances and has the advantage of being relatively unaffected by factors such as darkness, precipitation, fog etc. However, low resolution is a drawback of this system. Because resolution is determined by wavelength, systems using infrared radar with a shorter wavelength than radio waves also are in use. These provide better resolution, but their identification range is comparatively short, and they tend to be adversely affected by bad weather conditions. Since each approach has its advantages and disadvantages, monitoring system are adopting a combination of multiple sensors of different types.

For a robotic car capable of autonomous driving, accurate determination of the vehicle's position and of the peripheral situation is absolutely necessary. This means that such a car requires more capable "eyes" than what has been available so far. One possibility is laser radar, also known as LIDAR (Light Detection and Ranging). The principle is the same as a 3D laser scanner used for purposes such as surveying of structures, where a laser beam scans an object to measure the direction and distance of an object. In the case of a robotic car, such a system can be mounted for example on the roof to scan the periphery in a 360 degree arc as well as in the vertical direction. The point cloud data obtained in this way are processed by a computer to create a 3D map of the surroundings in real time, which is compared to GPS based position information, map data etc., thereby enabling autonomous driving.

This has brought the robotic car from a mere dream into the realm of reality, but there are still significant hurdles to be overcome. On public roads, unexpected situations may be encountered suddenly, such as disasters, changing weather conditions, and so on. The development of artificial intelligence (AI) that allows coping with such situations on a flexible case by case basis is therefore still required.