Object recognizing apparatus
View Patent ↗An existence probability of an image object based on an image taken by a stereo camera 2 is calculated by an image object existence probability calculating portion 11 , an existence probability of a millimeter wave object based on an output of a millimeter wave radar 3 is calculated by a millimeter wave object existence probability calculating portion 12 , and an existence probability of a laser object based on an output of a laser radar 4 is calculated by a laser object existence probability calculating portion 13 . Further, the respective existence probabilities of the image object, the millimeter wave object, the laser object are corrected based on recognizing rates of the respective recognizing sensors by an existence probability correcting portion 14 , the existence probabilities after correction is fused as a fusion existence probability by a fusion existence probability calculating portion 15 , thereby, a control of avoiding contact with a hazard or alarming or the like is made to be able to execute by a firm and optimum timing.
1. An object recognizing apparatus for recognizing an object based on signals received from a plurality of sensors, comprising:
an existence probability calculating unit configured to calculate each existence probability of the object to each sensor based on a normal distribution centering on an output value of each sensor;
an existence probability correcting unit configured to calculate each corrected existence probability by correcting said each existence probability with a recognizing rate of each sensor; and
a fusion existence probability calculating unit configured to calculate a fusion existence probability of the object by fusing said each corrected existence probability.
2. The object recognizing apparatus according to claim 1 , wherein
the fusion existence probability is calculated based on a multiplied value calculated by multiplying each nonexistence probability of the object calculated based on the each corrected existence probability.
3. The object recognizing apparatus according to claim 1 , wherein
the fusion existence probability is calculated based on each multiplied value calculated by multiplying a reliability of each sensor and the each corrected existence probability.
4. The object recognizing apparatus according to claim 1 , wherein
said existence probability is calculated as a probability distribution after a set time period predicted from a current movement and a position of the object.
5. An object recognizing apparatus for recognizing an object based on signals received from a plurality of sensors, comprising:
an existence probability calculating unit configured to calculate each existence probability of the object to each sensor based on a normal distribution centering on an output value of each sensor;
an existence probability correcting unit configured to calculate corrected existence probability by correcting said each existence probability in accordance with a stabilization of own vehicle; and
a fusion existence probability calculating unit configured to calculate a fusion existence probability of the object by fusing said each corrected existence probability.
6. The object recognizing apparatus according to claim 5 , further comprising:
a vehicle stability control apparatus to stabilize running condition of own vehicle, wherein
the existence probability correcting unit corrects the each existence probability based on a characteristic of the vehicle stability control apparatus.
7. The object recognizing apparatus according to claim 6 , wherein
the vehicle stability control apparatus is at least one of an antilock-brake system and a vehicle dynamics control.
8. The object recognizing apparatus according to claim 5 , wherein
said existence probability is calculated as a probability distribution after a set time period predicted from a current movement and a position of the object.
9. The object recognizing apparatus according to claim 6 , wherein
said existence probability is calculated as a probability distribution after a set time period predicted from a current movement and a position of the object.
10. An object recognizing apparatus for recognizing an object based on signals received from a plurality of sensors, comprising:
an existence probability calculating unit configured to calculate each existence probability of the object to each sensor based on a normal distribution centering on an output value of each sensor;
an existence probability correcting unit configured to calculate corrected existence probability by correcting said each existence probability with each characteristic of sensors set in accordance with a running environment of own vehicle; and
a fusion existence probability calculating unit configured to calculate a fusion existence probability of the object by fusing said each corrected existence probability.