JP2007240384A - Radar device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a radar device keeping on estimating the existence position of an object falling outside a detection area and promptly detecting an obstacle, etc. intruding from outside the detection area into the detection area. <P>SOLUTION: This laser radar device (distance measuring device for a vehicle) is equipped with a scanner 13 performing two-dimensional scanning with a laser light and a control circuit 11. The control circuit 11 extracts single-slope distribution at both side ends of the detection area of the scanner 13. Interruption information is recorded consisting of distance values of the single-slope distribution, a received light level, a detection area width, and an area value. Interruption information extracted by a frame of this time is compared with interruption information in the past to calculate interruption possibility. An intruding vehicle from outside the detection area is promptly detected based on the interruption possibility. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a radar device that measures the presence of an object existing in front, the distance to the object, and the direction by scanning electromagnetic waves such as laser light and millimeter waves.

  2. Description of the Related Art Conventionally, there is a vehicle distance measuring device that scans the front of a vehicle with a laser beam or the like to measure whether or not an object exists and the distance to the object. Using such a vehicle distance measuring device, a distance between the preceding vehicle is measured, and a constant inter-vehicle distance following traveling (ACC: Adaptive Cruise Control) is performed to keep the inter-vehicle distance constant. In addition, safety control is performed such that a warning sound is emitted to the driver, the seat belt is tightened, and braking is performed when the distance from the obstacle becomes a safe distance or less.

  In such a vehicle range finder, a measurement operation is performed for each of a plurality of detection areas obtained by dividing a detection area (substantially the same range as a laser beam irradiation range) extending in the scan direction (scan direction) in the scan direction. It was. In recent years, in order to widen the application range of the safety control or the like to scenes such as general roads and congested roads, a wider range of detection is required in vehicle ranging devices. In order to expand the detection area, it is conceivable to expand the scan range. However, if the scan range is simply expanded, the cost increases and the size of the apparatus increases. Therefore, a radar apparatus has been proposed in which the beam width of the laser beam is increased at the end of the scan range and the detection area is expanded (see, for example, Patent Document 1).

In addition, an object detection device that estimates the position of the entire object based on the width of the detected object is proposed even when an already detected object moves and moves out of the detection area and the entire object cannot be detected. (See, for example, Patent Document 2).
JP 2005-283138 A JP 2002-296350 A

  However, if the detection area is enlarged unnecessarily as in the device of Patent Document 1, there is a problem that a vehicle traveling in the adjacent lane is erroneously detected. Here, in the apparatus of Patent Document 1, a configuration is disclosed in which a peak in the scan direction is extracted from the reception intensity data for each division detection area, and if no peak is detected, it is determined that no object exists. In such a configuration, an obstacle that newly enters the detection area from outside the detection area cannot be detected at an early stage. In other words, it is determined that an obstacle that suddenly approaches from outside the detection area (for example, a car that suddenly changes its lane from the adjacent lane to its own lane) is not an object until a peak in the scan direction is detected. Therefore, there is a risk that safety control such as braking is not in time.

  Further, in the apparatus of Patent Document 2, it is possible to continue to estimate the existence position of an object that falls outside the detection area from within the detection area. For this purpose, it is necessary to first obtain the width of the entire object, It was not possible to detect early obstacles that entered the detection area from outside the area.

  An object of the present invention is to provide a radar apparatus that can early detect an obstacle or the like that enters the detection area from outside the detection area.

  The radar apparatus according to the present invention includes an electromagnetic wave irradiation unit that irradiates an electromagnetic wave, an electromagnetic wave reception unit that receives a reflected wave from the irradiation direction of the electromagnetic wave, an electromagnetic wave irradiation direction of the electromagnetic wave irradiation unit, and the electromagnetic wave reception unit A scanning unit that oscillates the receiving direction of the electromagnetic wave in a range of a predetermined detection area, and the electromagnetic wave irradiating unit irradiates the electromagnetic wave irradiating unit with an electromagnetic wave for each irradiation direction in angular increments due to the swaying, and the electromagnetic wave receiving unit And a control unit that measures the distance of the reflection point in the irradiation direction based on the peak on the time axis of the received wave intensity of the reflected wave. A determination process for determining that there is an object at the distance in which the reflected wave intensity at the reflection point is maximum, and a direction in which the reception intensity of the reflection point is maximum when there are reflection points at substantially the same distance in each of a plurality of adjacent irradiation directions; Multiple at the end of When there is a reflection point of approximately the same distance in each of the irradiation directions and the received intensity of the reflection point is rising toward the end of the detection area, an object exists outside the detection area. An estimation process for estimation is executed.

  In this invention, in each irradiation direction, the distance value of the reflection point (target) is detected based on the peak of reflection intensity on the time axis (timing at which the maximum intensity is reached). Further, the distribution of the received wave intensity of each target in the scan direction is calculated. In the scan direction distribution of the received wave intensity, there is a maximum peak, and if the distance between the targets is the same, it is determined that an object is present. The scan direction distribution of the received wave intensity rises toward the end of the reflected wave detection area range, there is no peak where the received wave intensity becomes maximum, and the distance between each target is approximately the same. It is estimated that there is an object.

  In the radar apparatus of the present invention, when the control unit determines that the object exists a plurality of times in substantially the same direction and the same distance in the determination process, the object is in the direction and distance of the determination process. After determining that the object is present in the estimation process, the determination process determines that the object exists at substantially the same distance as the reflection point in the estimation process. When it is determined that the number of times is less than the number of times, it is determined that an object exists in the direction and distance of the determination process.

  In the present invention, when the distance value when it is determined that the object exists substantially matches the distance value when it is estimated that the object exists in the past, they are associated with each other. If the number of times determined to be an object is equal to or greater than the predetermined number, the object is determined, but if the object is associated with the distance value of the object estimated in the past, the number is less than that (for example, once) If it is determined to be an object, it is determined to be an object. When it is determined that the object is an object, it is notified to the host system (vehicle control device), and as a result, safety control is performed early.

  Further, in the radar apparatus of the present invention, the control unit further includes a reflection point having a predetermined intensity or more when the distance between the reflection points in the estimation process is substantially the same as the reflection point in the past estimation process. The width in the scanning direction that exists, the integrated value of the received wave intensity at each reflection point in that width, the width in the scanning direction in which there is a reflection point that is greater than or equal to the predetermined intensity in the past estimation process, and the reflection point in that width Calculate the possibility of interruption of the object estimated by comparing the integrated value of the received wave intensity, and if the possibility of interruption is equal to or greater than a predetermined value, there is an object near the end of the detection area It is characterized by determining.

  In the present invention, when it is estimated that an object exists, the target width from the detection area end (width in the scanning direction) based on the distance value and the received wave intensity of each target, and the integrated value of the received wave intensity in that width (Area value) is recorded as interrupt information. Past interrupt information and current interrupt information are associated with each other, and the interrupt possibility is calculated. For example, when each of the above values (target width, area value) increases, the possibility of interruption is increased. When the interrupt possibility becomes a predetermined value or more, it is determined that an object exists in the vicinity of the end of the detection area.

  Further, in the radar apparatus according to the invention, the control unit estimates the estimation process when the reflection point distance in the estimation process is substantially the same as the distance of the object determined in the past determination process. The presence position of the detected object is estimated based on the position of the object determined in the past determination process.

  In the present invention, when the distance value of the interrupt information substantially matches the distance value of the object when it is determined that the object exists in the past, they are associated with each other. Based on the past object information (distance, position, width, etc.), the estimated distance, position, width, etc. of the object are estimated.

  According to the present invention, the distribution in the scanning direction of the received wave intensity is calculated, the distribution shape of the object existing outside the detection area is extracted from the distribution at the end of the detection area, and the detection area enters the detection area from outside the detection area. It is possible to detect obstacles coming soon. Further, by recording the position of the detected object once, it is possible to continue to estimate the position of the object that is outside the detection area.

FIG. 1 is a block diagram of a laser radar device (vehicle ranging device) according to an embodiment of the present invention. FIG. 2A is a diagram illustrating an on-vehicle mounting example of a laser radar device, and FIG. 2B is a diagram illustrating a detection area of the laser radar device. The laser radar device 2 is installed so as to irradiate a laser beam in front of the own vehicle 1.
An LD (Laser Diode) drive circuit 10 controls light emission of the LD 12 based on the drive signal generated by the control circuit 11. The scanner 13 scans (scans) the laser light generated by the LD 12 within a predetermined scan range based on the control of the control circuit 11. The laser beam emitted from the scanner 13 is emitted in the traveling direction (forward) of the host vehicle 1 through the light projecting lens (see FIG. 2A). The scanning position detection device 14 detects the scanning position of the laser beam in the scanner 13 and outputs it to the control circuit 11.

The reflected light returned from the laser beam emitted from the LD 12 after being reflected by a front object (for example, a vehicle) as a detection target is collected by a light receiving lens, received by a PD (Photo Diode) 15, and reflected therefrom. A signal corresponding to the light intensity is output to the light receiving circuit 16. The light receiving circuit 16 digitizes the signal level of the input reflected light (hereinafter referred to as the light receiving level or the amount of received light) and outputs it to the control circuit 11. The control circuit 11 stores the input light reception level in the memory 17 corresponding to the scan position input from the scan position detector 14. The light emitted from the LD 12 is shaped into a beam having a predetermined divergence angle by the light projecting lens. The light projection intensity decreases as the distance from the beam center increases. In the present embodiment, the beam divergence angle is an angle at which the intensity becomes approximately half (or a predetermined ratio) with respect to the peak value of the emission intensity. This is because even if the light is extremely weak, there is reflected light from the object, but the light emission intensity is required to such an extent that the reflected light can be observed. For this reason, the beam divergence angle is a range that is half the light projection intensity at the center of the beam. As a result, since the projected light exists even outside the beam divergence angle, the reflected light may be observed when an object exists outside the defined beam divergence angle. . As shown in FIG. 2B, such a measurement is performed in a plurality of detection regions (laser irradiation directions) obtained by dividing the detection area (within the beam divergence angle) into a horizontal scan direction (scan direction) with a constant width. Do each. In each detection region, light emission and light reception are performed a predetermined number of times. As shown in FIG. 2B, the “detection area” refers to a predetermined range that is substantially the same as the laser irradiation range or narrower than the laser irradiation range. Usually, the center of the detection area (optical axis) is installed and used so as to coincide with the traveling direction when the vehicle 1 is traveling straight ahead.
In addition to the temporary variables used for the measurement, the memory 17 also stores programs such as a procedure for determining the detected object.

Next, FIG. 3 shows a configuration of a portion that supports the light projecting lens and the light receiving lens of the scanner 13.
In FIG. 3, a control signal from the control circuit 11 is input to the drive circuit 30. The drive circuit 30 supplies a drive current to the drive coil 31 based on the input control signal. The driving coil 31 generates a magnetic field according to the current value, and generates an attractive force or a reaction force with a permanent magnet (not shown) that is installed, so that the light projecting lens 33 and the light receiving lens 34 are integrally supported. The supporting members (not shown) to be moved are respectively moved in the horizontal direction. The support member is supported by the leaf spring 32 so as to be movable in the horizontal direction. Therefore, the support members (the light projecting lens 33 and the light receiving lens 34) move to a horizontal position where the force generated in the drive coil 31 by the drive current and the reaction force generated in the leaf spring 32 are balanced, and are stopped. The position of each lens is detected by a sensor (not shown) and the sensor output is input to the drive circuit 30 to constitute a servo mechanism.

  In this way, the light projecting lens 33 and the light receiving lens 34 can be moved to predetermined positions in the horizontal direction.

  FIG. 4 shows optical paths of the light projecting lens 33 and the light receiving lens 34 driven by the scanner 13. The light projecting lens 33 is provided on the front surface of the LD 12, and the light receiving lens 34 is provided on the front surface of the PD 15.

  The laser light emitted from the LD 12 is polarized in the center direction of the light projecting lens 33. When the position of the projection lens 33 is at the center, the laser beam is emitted to the front along the optical path as indicated by the solid line in FIG. The emitted laser light is reflected by a front object (for example, a vehicle), enters the light receiving lens 34 through an optical path as indicated by a solid line in FIG. 4, and is received by the PD 15.

  Further, when the projection lens 33 is moved upward in the figure by the scanner 13, the laser light is emitted upward in the figure along an optical path as indicated by a dotted line in FIG. The emitted laser light is reflected by an object in the upward direction in the figure, enters the light receiving lens 34 through an optical path as indicated by a dotted line in FIG. 4, and is received by the PD 15.

  In this manner, the scanner 13 scans the laser light in the horizontal direction by moving the light projecting lens 33 and the light receiving lens 34 integrally to a predetermined position in the horizontal direction. In the above example, a configuration is shown in which scanning is performed only in the horizontal direction. However, scanning may be performed in the vertical direction. When scanning in the vertical direction, the vertical angle may be changed at the end of the horizontal scan.

  The laser radar apparatus 2 according to the present embodiment performs the following processing with one horizontal measurement as one frame.

  The control circuit 11 detects a peak on the time axis from the light reception level (light reception amount) of each detection area recorded in the memory 17, and receives the peak as the light reception timing (time from light projection to light reception). Detect the amount. The control circuit 11 calculates the distribution of the received light amount in the scan direction. FIG. 5A shows an example of the distribution (graph) of the received light amount in the scan direction. The horizontal axis of the graph shown in FIG. 5A represents the scan horizontal direction, and the vertical axis represents the amount of received light. FIG. 5B shows a distribution representing distance values calculated based on the light reception timing for each detection region.

  The control circuit 11 detects a peak in the scan direction from the received light amount distribution shown in FIG. For example, a difference in the amount of light received between adjacent detection areas is calculated, and a peak is detected based on whether or not there is a portion where the difference changes from plus to minus with respect to a change in the detection area (scan direction). Further, by connecting the points where the respective received light amount data are plotted at the center position of each detection area with a straight line, a continuous line-shaped continuous waveform (see FIG. 3A) is set, and a peak is extracted based on this. It may be. When this peak is detected, the control circuit 11 determines that an object has been detected. Further, the width of the object is detected by determining both ends of the detection region (or the continuous waveform) where the light reception level is equal to or higher than a predetermined threshold as both ends of the object. This threshold value may be determined in advance, or may be determined from the peak received light level (for example, 50%).

  Further, the control circuit 11 determines whether or not there is an object entering from outside the detection area at each of the left and right ends of the detection area. FIG. 6A is a diagram illustrating an example in which an object (vehicle) exists outside the right end of the detection area. Thus, when an object is present in the vicinity even outside the detection area, the received light level rises to the left and right edges and does not reach the peak as shown in FIG. (There is no part that changes from negative to negative). When the control circuit 11 detects such a distribution of light reception levels, the light reception level of each detection region, the width of the detection region (width in the scanning direction) that is equal to or greater than the threshold value, and the area value of the amount of light received at each of the left and right ends. Information of (integral value of continuous waveform) is extracted as interrupt information.

  Further, as described above, the control circuit 11 determines the reflection point (referred to as a target) in each detection region from the own vehicle based on the time from when the laser light is emitted until the reflected light is received. Measure distance. FIG. 5B shows an example of the distribution of distances. The horizontal axis of this graph represents the scan horizontal direction, and corresponds to the scan direction distribution of the received light level shown in FIG. As shown in FIG. 5B, in this example, approximately the same distance value is obtained in each detection region. The control circuit 11 groups these based on the distance value of each target and the position in the scanning direction (laser irradiation direction), and determines that each target is the same object (or object candidate). In this embodiment, a configuration in which scanning is performed only in the horizontal direction is shown. However, when scanning in the vertical direction is further performed, based on three-dimensional values of a distance value, a horizontal detection region, and a vertical detection region. Grouping.

  The control circuit 11 calculates the movement vector (change in the existing position) of the object by performing the above processing every frame, and based on the width and reflection intensity of the object, whether the object is a preceding vehicle Determine whether or not. Further, it is determined based on the interrupt information whether there is an object to be interrupted in the detection area.

Hereinafter, specific processing of the control circuit 11 will be described in detail with reference to FIGS.
FIG. 7 is a diagram illustrating the overall processing flow of the control circuit 11.
When the process is started, first, in ST1, the above-described measurement operation (emission and reception of laser light and data acquisition of the received light level) is performed for each detection region. In ST2, it is determined whether or not the measurement operation has been completed for all detection regions. That is, it is determined whether or not data for one frame has been acquired. If the measurement operation has been completed in all the detection regions, the processes of ST3 to ST7 are sequentially executed. If the measurement operation has not been completed in all the detection regions, the process returns to ST1 and the measurement operation is repeated for the next detection region.

  In ST3, the targets are grouped and collected as a target set. As described above, when the distance value of each target and the position in the scan direction are close to each other, the control circuit 11 groups them. In ST4, individual objects and interrupt information are extracted (separated) from the targets grouped in ST3. In the extraction of the object, processing for calculating the center position (distance and direction) and width is executed. If there is no grouped target, this object extraction process may not be executed in order to simplify the process. In addition, this object extraction process may be executed only for the grouped target ranges instead of executing this object extraction process for the entire range of the detection area.

  Next, in ST5, the data of the object extracted in ST4 in the previous frame, the data of the object extracted in ST4 in the current frame, and the data of interrupt information are compared, and a process of associating them is performed. For example, a position range having a predetermined spread is set with reference to the position of the previous object, and if the position of the current object is within this position range, a process of determining the same object is executed.

  In ST6, the interrupt information data extracted in ST4 in the previous frame and the interrupt information data extracted in ST4 in the current frame are compared and correlated to calculate the interrupt possibility. Information that is highly likely to be interrupted is determined to be an object. Details will be described later with reference to FIG.

  In ST7, the detected object is recognized and notified to a host system (a vehicle control device or the like that controls the vehicle speed). Here, the recognition of the object is a process for determining that the detected object is an object when the detected object is continuously detected in ST5 and m frames (for example, m = 5) are continuously detected. .

The object extraction, interrupt information extraction process in ST4, the object information association process in ST5, and the interrupt information association process in ST6 will be described in detail below.
FIG. 8 is a flowchart showing the object extraction / interrupt information extraction process of ST4. First, in s11, the control circuit 11 calculates the light reception level distribution of the detection area from the light reception level of each detection region (see FIGS. 5A and 6B). After creating the distribution of received light levels, the presence or absence of a peak is determined (s12). For example, the above-described continuous waveform is differentiated, and the peak at which the differential value changes from positive to negative (the point at which the differential value becomes zero and becomes maximum) may be taken as a peak. If there is a peak, the center position and width are calculated assuming that an object exists, and this is recorded in the memory 17 as object information (s13). The center position may be determined based on the extracted peak data, and the width of the object may be determined from the width in the scanning direction of the received light level that is equal to or greater than the threshold.

  Next, at the left and right ends of the detection area, it is determined whether or not there is a distribution in which the light reception level increases toward the left and right ends and does not reach the peak (hereinafter referred to as a single slope distribution). It is determined whether or not they are equal, and it is determined whether or not there is an object in the vicinity of the detection area (s14). Although there is a possibility that it is a guard rail or the like on the side surface of the road only by the single slope distribution, an interrupted object can be extracted by using the distance value in each detection area as a criterion.

  FIG. 9 is a diagram showing the received light amount level when the laser beam is reflected on the guard rail. As shown in FIG. 4A, when a guard rail is present at the right end of the detection area and the road shape is a left curve, the right end of the detection area is closest to the vehicle, so the reflection intensity of the laser light is strong. A single slope distribution as shown in (B) is formed. However, even if such a single slope distribution exists, the distance distribution differs between the interrupting object and the guard rail as shown in FIG. FIG. 4A shows the distance distribution of the interrupting object, and FIG. 4B shows the distance distribution of the guard rail. In the case of an interrupting object, the target detected in each detection region shows substantially the same distance value, but in the case of a guardrail as shown in FIG. 9A, the distance value decreases toward the right end of the detection area. Therefore, the control circuit 11 determines that this object is an interrupted object when the single slope distribution exists and the distance values in the respective detection areas are substantially equal.

  If it is determined in s14 that a single slope distribution exists and the distance values in each detection area are substantially equal, the number of detection areas that are equal to or greater than the threshold (that is, the width in the scanning direction), the light reception level of each detection area, and the distance between each area The value and the area value of the single slope distribution are extracted and recorded as interrupt information in the memory 17 (s15). When the above process is completed, the process returns to the overall process flow of FIG.

  FIG. 11 is a flowchart showing the object information association process of ST5. In s21, each object extracted in the current frame and the interrupt information are compared with the object information of the past frame. As a result, first, in s22, it is determined whether there is past object information corresponding to the object information extracted in the current frame. For example, a position range having a predetermined spread may be set based on the object position in the previous frame, and it may be determined whether or not the object position of the current frame exists within this position range.

  If it is determined that there is corresponding past object information, these are associated, and the number of frames determined to be present is counted (s23). This number of frames is used for the object determination process in the process of ST7.

  FIG. 12 is a flowchart showing the interrupt information association process in ST6. In s31, each object extracted in the current frame and the interrupt information are compared with interrupt information of a past frame (hereinafter referred to as past interrupt information). As a result, first, in s32, it is determined whether there is past interrupt information corresponding to the interrupt information extracted in the current frame. For example, if the distance value in each region is substantially the same, a process of associating is executed.

If it is determined that there is corresponding interrupt information, the possibility of interrupt is calculated (s33). This interruption possibility represents the possibility that the single slope distribution detected at the left and right ends of the detection area is an object. It can be determined that the higher the possibility of interruption, the higher the possibility that an object will enter from outside the detection area. The interrupt possibility is expressed by the following formula.
Interruptability = previous value + Ka * ΔD + Kb * ΔE + Kc * ΔF
Here, Ka, Kb, and Kc are weighting coefficients for each term. ΔD is the amount of change in the received light level, which is a value obtained by subtracting the received light level of the past interrupt information from the received light level of the current interrupt information at the detection area end. ΔE is a change amount of the detection area width, and is a value obtained by subtracting the detection area width of the past interrupt information from the detection area width of the current interrupt information. ΔF is a change amount of the area value, which is a value obtained by subtracting the area value of the past interrupt information from the area value of the interrupt information of this time. Note that the initial value of the interrupt possibility (when there is no previous value) is zero if the interrupt information is not associated with past object information. The case where it is associated with a past object, that is, the interrupt information newly detected immediately after the disappearance of the peak will be described later.

  Next, the control circuit 11 determines whether or not this interrupt possibility has become a predetermined value or more (s34). If it is equal to or greater than the predetermined value, it is determined that the interrupt information is an object (s35). The presence direction of the object may be the end of the detection area. In this case, the object determination process of ST7 may be performed immediately and the object may be immediately determined and notified to the host system, or the object may be determined to be the object when the frame continues in ST7. May be notified. Further, as will be described later, an interrupt flag may be assigned to this object, and if it is continued for n frames (n <m) fewer than m frames in ST7, it is determined that the object is an object and notified to the host system. .

  Next, the control circuit 11 determines whether there is past interrupt information corresponding to a newly detected object (an object that has detected a peak but could not be associated with a past object in ST5) (s36). In this case, if the distance value in each area of the object and the distance value in each area of the past interrupt information are substantially the same, a process of associating is executed. If it is determined that there is corresponding interrupt information, an interrupt flag is assigned to this detected object, and the number of frames determined to be an object in ST7 is set to n frames (n <m, for example, n = 1) smaller than m frames. .

  Next, the control circuit 11 determines whether there is past object information corresponding to the current interrupt information (s38). In this case, if the distance value in each area of the interrupt information and the distance value in each area of the past object information are substantially the same, a process of associating these is executed. When the interrupt information is associated with past object information, the interrupt information is determined to be an object that has moved out of the detection area from the detection area, and based on the object information while retaining the past object information. Then, the position of the object that has moved out of the detection area is estimated (s39). The position of the object that has moved out of the detection area is the position of the object estimated from past object information and interrupt information. Note that the width of an object that has moved out of the detection area may be estimated from past object information. Any method for estimating the position of the object may be used. For example, the object position is estimated as follows.

First, the angle from the edge of the detection area is estimated based on the possibility of interruption. FIG. 13A is a diagram showing the angle from the detection area edge and the possibility of interruption. The horizontal axis of the graph shown in FIG. 5A indicates the angle from the detection area end, and the vertical axis indicates the value obtained by dividing the interrupt possibility by the interrupt possibility immediately after the peak disappearance. The interrupt possibility immediately after the peak disappearance represents an initial value of interrupt possibility when new interrupt information is extracted and associated with past object information. The possibility of interruption immediately after the peak disappearance is expressed by the following formula.
Interruptability immediately after peak disappearance = Ka * ΔG + Kb * ΔH + Kc * ΔI
Here, Ka, Kb, and Kc are weighting coefficients for each term. ΔG is the light reception level of the newly extracted interrupt information, and is the value of the light reception level at the end of the detection area. ΔE is the detection area width. ΔF is an area value. When the object moves out of the detection area and the peak disappears from the distribution of the received light level, new interrupt information is extracted, but by setting the initial value of the interrupt possibility as described above, it moves away from the end of the detection area. As a result, the possibility of interruption decreases. Thereby, the angle from the detection area end of the object is estimated. About the distance of an object, it estimates from the distance of the past object information matched. Alternatively, it may be estimated from the distance value of the object of the interrupt information extracted this time. The position of the object that has moved out of the detection area is estimated from the angle and distance value from the end of the detection area. The width of the object that has moved out of the detection area may be the width of the past object information.

  Alternatively, the front of the vehicle may be photographed using a camera (preferably a CMOS sensor having a high dynamic range) having a field of view wider than the detection area, and the position of the object may be measured using this image. What is necessary is just to use the position of the object measured by this sensor as the object information which went out of the detection area.

  In FIG. 12, finally, interrupt information and object information that are no longer necessary are deleted from the memory (s40). For example, the object information associated with the frame and the information for which there was no interrupt information are deleted. When these processes are completed, the process returns to the overall process of FIG.

  In ST7 of FIG. 7, the detected object is recognized and notified to the host system. Here, when the detected object is detected continuously for m frames using the count number in s23, this detected object is determined to be an object and notified to the host system. If the interrupt flag is assigned in s35 or s37, when the detection is continued for n frames, the detected object is determined to be an object and notified to the host system.

  As described above, the laser radar apparatus according to the present embodiment can detect an obstacle entering the detection area from the outside of the detection area at an early stage by using the interrupt information and the interrupt possibility. In addition, by associating the interrupt information with an object that has been detected in the past, it is possible to continue estimating the existence position of the object that is outside the detection area.

  Although the laser radar apparatus has been described in the present embodiment, the configuration of the present invention can also be used for other radar apparatuses such as a millimeter wave radar.

The block diagram of the laser radar apparatus which is embodiment of this invention Diagram showing an example of on-board mounting of a laser radar device and detection area The figure which shows the structure of the part which supports the light projection lens and light reception lens of a scanner. The figure which shows the optical path of the light projection lens and light reception lens which were driven by the scanner Diagram showing the distribution of received light level and distance value The figure which shows the example where the object exists in the detection area right neighborhood outside the detection area Flow chart showing the procedure of overall processing Flow chart showing specific procedures for object extraction and interrupt information extraction processing The figure which shows the light reception level when the laser beam is reflected on the guard rail Diagram showing distance distribution between interrupting object and guardrail Flow chart showing specific procedure of interrupt information association processing Flow chart showing interrupt information association processing Diagram showing angle from detection area edge and interrupt possibility

Claims (4)

An electromagnetic wave irradiation unit for irradiating electromagnetic waves;
An electromagnetic wave receiving unit for receiving a reflected wave from the irradiation direction of the electromagnetic wave;
A scanning unit that swings the electromagnetic wave irradiation direction of the electromagnetic wave irradiation unit and the reception direction of the electromagnetic wave reception unit within a range of a predetermined detection area;
For each irradiation direction of the angular increments due to the oscillation, the electromagnetic wave irradiation unit is irradiated with electromagnetic waves and the reflected waves are received by the electromagnetic wave reception unit, and the peak of the reflected wave on the time axis is received. A control unit for measuring the distance of the reflection point in the irradiation direction based on
A radar apparatus comprising:
The control unit determines, when there is a reflection point of approximately the same distance in each of a plurality of adjacent irradiation directions, a direction in which the received intensity of the reflection point is maximum, and that an object exists at the distance. ,
When each of a plurality of irradiation directions at the end of the detection area has a reflection point of substantially the same distance, and the received wave intensity at the reflection point increases toward the end of the detection area, A radar apparatus that executes an estimation process for estimating that an object exists outside a detection area. When it is determined in the determination process that an object is present a plurality of times in substantially the same direction and the same distance, the control unit executes a determination process for determining that an object exists in the direction and distance of the determination process,
After it is estimated that the object is present in the estimation process, and in the determination process, it is determined that the object exists at substantially the same distance as the reflection point of the estimation process, and the number of determinations is less than the determination number of the determination process. The radar apparatus according to claim 1, wherein it is determined that an object exists in the direction and distance of the determination process. When the distance between each reflection point in the estimation process is substantially the same as the reflection point in the past estimation process,
The width in the scanning direction where there is a reflection point above a predetermined intensity, the integrated value of the received wave intensity at each reflection point in that width, the width in the scanning direction where there is a reflection point above the predetermined intensity in the past estimation process, and Calculate the possibility of interruption of the object estimated by comparing the integrated value of the received intensity of each reflection point in that width,
The radar apparatus according to claim 1, wherein when the interrupt possibility becomes equal to or greater than a predetermined value, it is determined that an object is present in the vicinity of an end of the detection area. When the distance of the reflection point in the estimation process is substantially the same as the distance of the object determined in the past determination process,
The radar apparatus according to claim 1, wherein the presence position of the object estimated in the estimation process is estimated based on the position of the object determined in the past determination process.

Images