Method for determining object classification from...

Communications: directive radio wave systems and devices (e.g. – Return signal controls external device – Radar mounted on and controls land vehicle

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C701S301000, C180S169000, C340S436000

Reexamination Certificate

active

06680689

ABSTRACT:

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
Not Applicable.
BACKGROUND OF THE INVENTION
The present invention relates in general to side object detection systems for motor vehicles, and, more specifically, to classifying a type of object detected by a remote sensor within a detection zone to a side of a vehicle.
Automotive systems known as side object detection systems (SODS) utilize “side-looking” remote sensors for such applications as blind-spot detection and lane change aid. These applications are designed to alert the driver to potential hazards, e.g., objects that may be located adjacent to the host vehicle. The remote sensors may employ radar transceivers, light sensors, ultrasonic sensors, and other technologies.
One objective of the side-looking sensors is to identify the presence and location of objects within a predetermined zone of interest adjacent to the vehicle. Radar sensors detect and locate objects by transmitting electromagnetic energy which reflects off of objects within the sensor field-of-view. The reflected signal returns to the radar sensor where it is processed to determine the round-trip travel time of the transmitted/received energy. The round trip travel time is directly proportional to the range of the target from the radar sensor. In addition to range determination, there are methods to determine azimuth (i.e. cross-range) location of detected objects such as multiple scanned/switched beams and mono-pulse implementations. Therefore, depending upon its complexity, the radar is capable of locating objects in both range and azimuth relative to the sensor location.
Based upon the reflected signals during a sampling of the entire field-of-view, a set of detection points is accumulated. Due to the nature of “reflections” collected by a remote sensor (whether a radar, laser, ultrasonic, or other active sensor), the set of detection points is representative of only certain spots on the object or objects present in the sensor's field-of-view. The detection points must be analyzed in order to determine what type of objects may be present and where such object is located.
Based on the type (i.e., class) and location of detected objects, a blind spot detection or lane change aid system must decide whether a detection is one for which it should alert the driver. Under certain conditions, it may be undesirable to always generate an alert every time that any object is detected in the detection zone. For example, side-looking radar sensors will be subjected to reflections from common roadway structures such as guard-rails and roadside signs. These objects may not constitute a threat to which the driver desires to be alerted since they are stationary. However, due to the complexity of the driving environment, it has not been possible for a radar sensor to discriminate between the various driving scenarios without extensive processing and expensive sensor designs. It would be desirable to discriminate between objects for which an alert should or should not be provided using relatively simple sensors and without excessive computational resources.
Occupant safety systems are known that include pre-crash functions based on sensing an impending collision and taking advance actions to increase protection of the vehicle occupants. Potential pre-crash actions include pre-tensioning of seatbelts and adapting the deployment of airbags in response to the expected point of impact. Prior art pre-crash systems have employed forward-looking or rearward-looking radar wherein velocity of an object includes a radial component thereby permitting the detection and localization of objects based on Doppler measurements. In side-looking systems, any radial velocity is small and Doppler measurements are impractical. Nevertheless, the ability to determine a side location and the type of object about to be impacted could be used to improve the pre-crash actions.
SUMMARY OF THE INVENTION
The present invention has the advantage of classifying a type of object within a zone of interest to the side of a transportation vehicle using relatively simple and inexpensive remote sensors and signal processing.
In one aspect of the invention, a method is provided for classifying an object type of a 3-D object sensed by a remote sensor for detecting objects to a side of a transportation vehicle. The transportation vehicle moves along a front-to-rear directional axis. The remote sensor is mounted at a predetermined reference point on the transportation vehicle. A set of detection points is identified substantially to the side of the transportation vehicle using the remote sensor. A closest one of the detection points to the remote sensor is identified. If the closest one of the detection points is at a substantially perpendicular direction from the remote sensor, then a size of the object is determined in response to an area defined by locations of the set of detection points. The size of the object is compared with a size threshold. If the size is greater than the size threshold then the object is classified as a stationary object.


REFERENCES:
patent: 4257703 (1981-03-01), Goodrich
patent: 5325096 (1994-06-01), Pakett
patent: 5339075 (1994-08-01), Abst et al.
patent: 5517196 (1996-05-01), Pakett et al.
patent: 6243024 (2001-06-01), Yamabuchi et al.
patent: 6265968 (2001-07-01), Betzitza et al.
patent: 6396435 (2002-05-01), Fleischhauer et al.
patent: 6400308 (2002-06-01), Bell et al.
patent: 6498972 (2002-12-01), Rao et al.
patent: 6509863 (2003-01-01), Zoratti et al.
patent: 6542111 (2003-04-01), Wilson
patent: 2003/0025597 (2003-02-01), Schofield

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Method for determining object classification from... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method for determining object classification from..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining object classification from... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-3261146

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.