New Bosch pedestrian protection system helps drivers brake and take evasive action
Technology looks one second into the future
To test the technology, B?rkle and his interdisciplinary team have built a research vehicle. Its central component is a Bosch stereo video camera of the kind already used in production models. Mounted behind the windshield near the rear-view mirror, the camera provides a 3D image of the area to the front of the vehicle, and detects pedestrians and oncoming traffic as well as obstacles on the road ahead. A computer in the trunk of the research vehicle analyzes the information. If a pedestrian suddenly appears in the stereo video camera’s field of vision, the system computes the likelihood of a collision as well as the route that must be taken to avoid it. All this happens at lightning speed – more than ten times a second. The correct interpretation of the images from the camera and the specific driving situation is particularly complex. “To plan the new trajectory as precisely as possible, we have to do things such as predict where the pedestrian is likely to be in a second’s time,” B?rkle explains. The team’s work focuses on developing the algorithms this requires. Bosch’s multi-faceted software expertise, which the company continues to expand, is vital in this process.
Key competence for automated driving
With their work on the analysis of camera images, the Bosch researchers are also making an important contribution to the development of automated driving. From 2020, it is expected that the Bosch highway pilot will enable highly automated freeway driving without the need for constant driver supervision. Among other things, this automation will be based on various sensors that provide a precise image of the vehicle’s surroundings. Here, Bosch relies on its mid- and long-range radar sensors, on its stereo video camera, and on its image-processing expertise. Bosch’s main goal in developing automated driving is greater safety on the roads. An estimated 1.3 million people worldwide are killed in road accidents each year. Ninety percent of these accidents are caused by human error. In difficult or confusing traffic situations, machine support can save lives.
The ultimate goal, therefore, is automated driving. In the meantime, Bosch will launch a whole range of useful driver assistance systems. Image analysis and the ability to compute new trajectories could also be used in an assistance system that guides vehicles through tight spaces. Roads are often clogged by cars parked on both sides, especially in cities. Things can get extremely tight if a van double-parks to make deliveries. The images from the stereo video camera can then provide crucial information. The computer analyzes it, and the assistant controls the power steering to enable the car to pass by without mishap, even when there is little room. “The examples show how Bosch is using sensors, software, and expertise in image processing to make mobility safer,” says Dr. Michael Bolle, head of corporate research and advance engineering at Bosch.
A network spanning industry and academia
The emergency braking assistant, the evasive steering support for pedestrian protection, and the assistance system for tight spaces are being developed as part of the publicly funded UR:BAN project. UR:BAN brings together 31 partners from the automotive, automotive-supply, electronics, communications, and software industries, as well as universities, research institutes, and municipal authorities. The aim of the partnership is to develop driver assistance and traffic management systems for the urban environment. Financial support for the project is being provided by the German Federal Ministry for Economic Affairs and Energy. Bosch knows that close cooperation between business and the academic world is conducive to innovation. This is why the company works with almost 250 universities and research institutes worldwide.
Комментарии