Data processing: vehicles – navigation – and relative location – Navigation – Employing position determining equipment
Reexamination Certificate
2001-07-24
2003-07-22
Cuchlinski, Jr., William A. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Navigation
Employing position determining equipment
C701S213000, C701S214000
Reexamination Certificate
active
06597984
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a method for determining the position of a motor vehicle with respect to a traffic lane, as well as to a device which is suitable for carrying out the method.
To increase convenience and safety in vehicles, sensor systems for environmental sensing which feature an intelligent scene interpretation are increasingly gaining importance.
German patent document DE 19749086 C1 describes a system for acquiring data which are indicative of the course of a traffic lane, containing a traffic lane detection sensor system in the form of a video camera and an object position sensor system. In this context, the traffic lane detection sensor system is composed of a video camera and the object position sensor system is composed of a radar sensor which measures at least the distance from an object located ahead of the vehicle and its heading angle with respect to the vehicle's moving direction. Within a vehicle motion model, this sensor data is fused for deriving a position estimate of the vehicle therefrom.
Important additional information on the vehicle's surroundings can be acquired with the aid of navigation systems. In this context, the positional resolution is generally precise enough to correlate the vehicle to the existing digital road map. However, if the vehicle enters a transition or branch region where, for example, a side road branches from a major road, then it is difficult to determine whether the vehicle continues to move forward on the major road or whether it moves toward the branching road or side road. For improved scene interpretation, it is therefore required for the navigation data to be usefully supplemented by additional sensor data.
The method described in German Patent Application DE 19906614 A1 therefore compares the data of a video-based traffic lane detection sensor system which detects the marking lines at the road edge to the data of a digital road map which is in communication with a navigation device. This makes it possible to determine the position of the vehicle with respect to the marking lines.
On the other hand, Japanese publication JP10325869 introduces a system in which the traffic lane detection sensor system connected to a digital road map is composed of a radar sensor. First of all, this makes the system suitable in all weather conditions (with no significant restriction by fog) and, secondly, independent of the existence of boundary lines at the road edge. However, this method has the disadvantage that radar-reflection bodies must be installed at equal intervals at the road edge.
SUMMARY OF THE INVENTION
An object of the present invention is to find a method for determining the position of a motor vehicle with respect to the traffic lane in which the data from a digital road map coupled with a navigation system is used with data delivered by a distance-resolving sensor.
The present invention provides a method for determining the position of a motor vehicle with respect to a traffic lane, in which a first correlation of the motor vehicle to a road is carried out via a road map which is coupled with a navigation system. This first correlation is made more precise by subsequently determining the distance of the motor vehicle from the road edge via a distance-resolving sensor. The signature of the received signal of the distance-resolving sensor is evaluated for determining the distance from the road edge.
In this context, the signature of the distance-resolving sensor is evaluated for determining the distance from the road edge.
The advantage of the present invention ensues from the possibility for a vehicle to autonomously determine its instantaneous position with respect to the traffic lane without falling back on a stationary traffic infrastructure (marking lines, reflectors).
Within the scope of the method according to the present invention, a distance-resolving sensor, preferably a millimeter wave sensor (however, a radar sensor or a distance-resolving sensor on the basis of a laser, for example, LIDAR, is also conceivable) for determining the vehicle distance from the road edge is combined with a digital road map coupled with a navigation system for position determination. Suitable as navigation system is, in particular, a DGPS (global positioning) system as is already frequently used in modern motor vehicles. The method according to the present invention advantageously eliminates this lack of clarity by correlating the map data with the distance estimate of a distance-resolving sensor. In this manner, the lateral vehicle coordinates and the vehicle orientation can be clearly determined. This is also important, in particular, if the intention is to detect whether a vehicle driving ahead is traveling on one's own or on the adjacent lane or whether stationary obstacles are located on one's own traffic lane.
Advantageously, the method according to the present invention additionally may use the data of a gyroscopic system integrated in the vehicle to be able to better determine the orientation of the vehicle on the road. In lieu of a gyroscopic system, it is, of course, also conceivable to use the position data of other systems and sensor technology contained in the motor vehicle.
Used as estimate of the distance of the motor vehicle from the road edge is, in an inventive manner, the distance of the range window within the received signal of the distance-resolving sensor in which the signature of the received signal begins to change significantly. In this context, it is conceivable for a significant change to be identified when the intensity of the received signal increases monotonously. This is a reliable procedure since a road surface appears to be smooth to the electromagnetic waves emitted by the radar sensor and, therefore, reflects back relatively little energy to the sensor while the reflections at the generally inhomogeneous road edge (gravel, grass, road pavement/curb edge transition) are significantly stronger. To prevent errors in the interpretation of the received signals resulting, for example, from signal interference, it is also conceivable for a monotonous increase in the signal to be rated as significant only if the increase in the signal energy takes place over at least two range windows. A useful selection of the number of range windows ensues in connection with the instantaneous distance resolution of the sensor system. Correspondingly, it is also conceivable in an advantageous manner to compute a distance estimate with respect to the monotonous transition of the road pavement to the road edge using suitable estimation methods known from image processing, the distance estimate then being considerably more accurate in its distance resolution than the resolution of the distance-resolving sensor. Using such an edge estimate whose resolution lies in the subpixel range, it is possible to achieve an increase in resolution by a factor 5 to 10.
However, it is in particular also possible for the received signal of the distance-resolving sensor to be examined for the presence of signatures which are typical of a road edge. In this context, it is particularly advantageous for the sensor to be configured to be sensitive to polarization so that the polarization of the received signal can be examined within the scope of the signature analysis. Investigations have shown that the polarization signatures of road surfaces and of textures typically found at the road edge (grass, crushed stone) differ significantly.
Advantageously, the course of the road can be determined from the chronological sequence of the distance information. In this context, it is conceivable for the course of the road to be determined with the aid of the method of least error squares or to use a Kalman filter which is optimized for this task. In this context, it is particularly beneficial for the Kalman filter to be adapted to data as, for example, curve parameters which is obtained from the road map. In the event that the sensor is temporarily not able to detect a road edge, for example in the r
Appenrodt Nils
Berner Joerg
Mezger Klaus
Wanielik Gerd
Cuchlinski Jr. William A.
Daimler-Chrysler AG
Davidson Davidson & Kappel LLC
Marc-Coleman Marthe
LandOfFree
Multisensory correlation of traffic lanes does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Multisensory correlation of traffic lanes, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multisensory correlation of traffic lanes will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3016388