Denso Corporation

Nagoya University

Hiroaki Kumon

Yukimasa Tamatsu

Kazuma Natsume

Akira Isogai

Hiroshi Murase

Eiji Teramura

Hiroshi Hazumi

Kenta Hoki

Takayuki Miyahara

The vehicle detecting apparatus includes an image sensor mounted on a host vehicle so as to be able to take an image ahead of the host vehicle, a light source area extracting function of extracting, from image data outputted from the image sensor, an area having luminance higher than a predetermined value as a light source area, a vehicle detecting function of detecting existence of at least one of an oncoming vehicle and a preceding vehicle by recognizing which of a headlight of the oncoming vehicle, a taillight of the preceding vehicle, and a roadside reflector causes the light source area in the image data. The vehicle detecting function is configured to lower a probability that the vehicle detecting function recognizes that the light source area is caused by the roadside reflector when the headlight of the host vehicle is in a low-beam position.

Vehicle detecting apparatus

An image processing device for a vehicle includes a radar for transmitting a radio wave outside of the vehicle and detecting an object in a first area outside of the vehicle by using a reflected wave of the transmitted radio wave; a camera for acquiring an image in a second area including the first area; and an image processing unit for processing the acquired image to detect, in the acquired image, the object detected by the radar. Visibility outside of the vehicle is detected based on result of detection of the radar and result of detection of the camera.

Image processing device for vehicle

A vehicle detection apparatus detects vehicles traveling on a road with their lights turned on. The apparatus captures images of a road, detects a light spot from captured images, and detects a first vertical coordinate of the light spot in the images. The apparatus further calculates a distance to the light spot based on a horizontal length of the light spot in the images and detects a second vertical coordinate corresponding to the calculated distance, based on correlated data in which vertical coordinates and distances in an image are correlated in advance. The apparatus further calculates a pitching amount that indicates a difference between the first and second coordinates and determines whether or not the light spot originates from a vehicle, based on the pitching amount.

Apparatus and method for detecting vehicles by identifying light spots from captured images

An apparatus for detecting a direction of a target is provided. The apparatus transmits and receives radio waves through a plurality of transmission/reception channels causing a phase difference in signals to be received through the transmission/reception channels and calculates the direction based on the phase difference. The apparatus comprises a direction calculating device, range determining device, and direction correcting device. The direction calculating device calculates the direction of the target based on the phase difference in the received signals on the assumption that the phase difference is within a range of &#x2212;&#x3c0; to +&#x3c0; [rad]. The range determining device determines that the target exists in any of azimuthal angle ranges each corresponding to ranges defined by (2m&#x2212;1) &#x3c0; to (2m+1) &#x3c0; [rad] (m is an integer). The direction correcting device corrects the direction according to a range-determined result.

Apparatus for detecting direction of target using difference in phase of radio wave signals received through plural channels

An average power value of a peak pair corresponding to a subjective target objective is converted into a radar cross section, to calculate a normalized average power value NP and a standard deviation DP representing a temporal dispersion of a power difference between peak pairs. When the value NP is larger than an automotive vehicle discriminating threshold THnp, the attribute of the subjective target objective is set to 'automotive vehicle.' When the value NP is not larger than the threshold THnp and the deviation DP is larger than a human objective discriminating threshold THdp, the attribute of the subjective target objective is set to 'non-vehicle objective: human objective.' Furthermore, when the value NP is not larger than the threshold THnp and the deviation DP is not larger than the threshold THdp, the attribute of the subjective target objective is set to &#x201c;non-vehicle objective: non-human.&#x201d;

Method and apparatus for discriminating a target objective, and related program

Provided is a radar apparatus capable of continuously and stably making a detection of a target even if a reflected wave from a target already detected falls into obscurity due to the presence of low-frequency noises or reflected waves from other targets. An estimated value of information on a target to be obtained when the target is detected in the present cycle are acquired from the target detected in a previous cycle. When a peak compatible with the estimated value is detected in only one of a frequency-rising section and a frequency falling section of a radar wave, if the frequency of the non-detected peak pertains a low-frequency noise domain or if a side-by-side travel flag is set with respect to the target detected in the previous cycle, the non-detected peak is considered as buried by low-frequency noises or peaks of other targets, and a peak pair corresponding to the detected target is extrapolated.

Radar apparatus, and program therefor

A cruise control apparatus can control a host vehicle to run with a fixed separation from a preceding vehicle when such a preceding vehicle is detected based on received radar signals and to run at a preset fixed speed when no preceding vehicle is detected. The cruise control apparatus is configured to respond to one of a set of predetermined actions by the vehicle driver, which indicate an intention to change the vehicle speed, by making it easier or more difficult for a preceding vehicle to be detected, in accordance with whether the indicated intention may signify that an actual preceding vehicle is not being detected or may signify that a non-existent preceding vehicle is being detected and that the host vehicle speed has been reduced accordingly. Improved performance is thereby achieved by utilizing the cognitive abilities of the driver to augment the detection operation of the cruise control apparatus.

Cruise control apparatus performing automatic adjustment of object recognition processing in response to driver actions relating to vehicle speed alteration

A vehicle radar system extracts peak frequencies fbu and fbd of respective beat signals Bto Brepresenting the frequency difference between a transmission signal fs and a plurality of received signals frto fr. The phase difference of respective beat signals Bto Bat the peak frequencies fbu and fbd is converted into a frequency signal. In the case of reflection from a close range road surface or raindrops, the phase difference of each beat signal is irregular. The peak frequency intensity of a converted frequency signal is small. This system compares the peak frequency intensity of the converted frequency signal with predetermined criterion intensity. Then, the system identifies an objective with a close range road surface or raindrops when the peak frequency intensity of the converted frequency signal is not larger than the predetermined criterion intensity.

Vehicle radar apparatus

An inventive frequency modulated continuous wave (FMCW) radar system realizes both a quick detection of a higher relative speed provisional target and a sure detection of a smaller relative speed provisional target. The number of detection cycles used for a paring validity check, used to see if a detected target or a pair of frequencies is an actual target or a pair for an actual target, is set in response to the relative velocity enabling the target information for a target of higher relative velocity to be output more quickly.

FMCW radar system

A radar system mounted in a reference vehicle can detect a distance and orientation of a preceding vehicle to thereby compute a relative position of a width center of the preceding vehicle. A curving radius of the reference vehicle is then detected for computing a relative rotation angle between a direction from the reference vehicle and a longitudinal direction of the preceding vehicle. Relationship between a relative rotation angle and a lateral bias of the relative position of the width center is previously prepared in a map. The computed relative rotation angle is applied on the map, so that the corresponding lateral bias is obtained to correct the computed relative position of the width center of the preceding vehicle. Thus, the width center of the preceding vehicle moving in an adjacent lane can be accurately estimated.

In-vehicle radar system

A vehicle-mounted radar apparatus which periodically derives and registers successive momentary position values for a target object such as a preceding vehicle based on received reflected radio waves and derives final lateral position data by smoothing the momentary position data, judges when a degree of scattering of the registered momentary position values exceeds a first predetermined level and in that case derives corrected position data based on differences between envelope curve line values which are generated based on local extreme values of the momentary position data, and performs smoothing of the corrected position data instead of the momentary position data, to obtain the final lateral position data. If the target object is not estimated to be located directly ahead of the host vehicle along a straight route, the corrected position data are adjusted in accordance with relative positions and orientations of the target object and host vehicle.

Vehicle-mounted radar apparatus providing improved accuracy of detection of lateral position of preceding vehicle

Signal process apparatus for an on-vehicle radar and method thereof

The invention provides an object recognition apparatus that removes virtual images without detecting a roadside object. When a predetermined vehicle is currently running on a lane, a detection area, outside of the currently running lane, is designated to include a first area and an assumed ghost area therein. The first area is enclosed by a boundary of the detection area, a boundary of the assumed ghost area adjacent thereto, and a boundary on the currently running lane whose distance is traveled by the vehicle during one control cycle. The invention detects an object for the first time in the assumed ghost area, not in the first area. When the object travels with a distance and a speed of a target vehicle, it is determined to be a ghost to be deleted.

Object recognition apparatus and method thereof

A radar apparatus is provided which is capable of discriminating between azimuth angles of two or more targets moving side by side close to each other. The radar apparatus provides antenna beams which partially overlap with each other to define a plurality of monopulse areas and processes input signals produced in each of the monopulse areas to obtain angular direction data. The radar apparatus determines a time-sequential variation in angular direction data in each of the monopulse areas and determines the angular direction data whose variation is within a preselected allowable range as being effective in determining the angular direction of each target correctly.

Monopulse radar apparatus

An FM-CW radar is provided which may be employed in anti-collision systems or cruise control systems installed in moving objects such as automotive vehicles. The radar produces a first spectrum using a portion of a beat signal in a frequency rising range wherein the frequency of the radar wave increases and a second spectrum using a portion of the beat signal in a frequency falling range wherein the frequency of the radar wave decreases and moves the second spectrum by frequency shifts which are determined as a function of the speed of a radar-mounted vehicle and corrected for compensating various errors involved in measuring the speed of the vehicle to form spectrum groups each consisting of the frequency components moved by one of the corrected frequency shifts and frequency components in the other of the first and second spectra. The radar selects an optimum one of the spectrum groups for use in determining target data.

Growing community of inventors

Kariya, Japan

Hiroaki Kumon