JP2014211332A - Radar device and control method thereof - Google Patents

Abstract

A radar device capable of correctly tracking a target object even when a reflected wave from the target object interferes with a reflected wave from another object. A radar apparatus according to the present invention receives a signal at a first time point from a radar that observes a surrounding object in a first cycle, and detects detection position information of the object from the received signal. And estimated position information of the object at a second time point after the first cycle from the first time point from the observation position information detected by the detection unit, and the estimated position information and the estimated position Pairing the observation position information at the second time point within the information pairing range to estimate the estimated position information of the object at the third time point after the first cycle from the second time point. A tracking unit. [Selection] Figure 1

Description

  The present invention relates to a radar apparatus and a radar apparatus control method.

  Conventionally, a radar device has been used as an object detection device that detects a relative speed with an object existing in the vicinity. The radar apparatus can obtain information such as the position, size, and relative speed of the detected object from the reflected wave.

  As an in-vehicle radar device, for example, a UWB radar device equipped with a UWB radar has been proposed in order to widen the detectable range of an object. For example, the UWB radar device transmits a monopulse toward the periphery of the host vehicle at a predetermined time interval, compares the monopulse interval of the received reflected wave with the transmitted monopulse interval, and compares the position and relative position of the object around the host vehicle. Measure speed. The measured object is tracked over time to determine whether it is a stationary object (stationary object) such as a guardrail, utility pole, or parked vehicle, or another vehicle (moving object) such as a running bicycle. can do.

  However, the reflected waves are superimposed on the reflected waves from other vehicles and bicycles that you want to detect and the reflected waves from stationary objects such as guardrails and parked vehicles, and contain many detection errors. .

  In order to improve detection accuracy, for example, there has been a conventional technique in which detected targets are grouped for each speed range and tracking is performed for each speed group.

Further, there has been a conventional technique for identifying a fixed side road by comparing the speed of an object on the left side of the host vehicle and an object on the left front side of the host vehicle.
(For example, see Patent Document 1 and Patent Document 2)
JP 2011-158292 A JP 2007-292463 A

  However, in the prior art, when a measurement object with low reflected wave intensity, such as a bicycle, approaches an object with high reflected wave intensity, such as a guardrail, the reflected wave is superimposed due to the reflection wave. In some cases, the measurement position of a measurement object having a weak wave is greatly displaced in the direction of an object having a strong reflected wave.

  In addition, when the reflected waves from two objects are superimposed, the peak of the measurement target having a weak reflected wave disappears, and the measurement target may not be detected.

  These detection errors and non-detections of the detection target make it difficult to track an object to be detected.

  Therefore, an object of the present invention is to provide a radar apparatus that can correctly track an object even when a reflected wave from the object interferes with a reflected wave from another object.

  In view of the above problems, a radar apparatus according to the present invention receives a signal at a first time point from a radar that observes a surrounding object in a first cycle, and detects observation position information of the object from the received signal. The estimated position information of the object at the second time point after the first cycle is estimated from the first time point from the observation position information detected by the detection unit and the detection unit, and the estimated position information and the estimated value Estimating estimated position information of the object at a third time point after the first cycle from the second time point by pairing with observation position information at the second time point within the pairing range of the position information And a tracking unit.

  According to the embodiment of the present invention, it is possible to provide a radar device that can correctly track an object even when a reflected wave from the object interferes with a reflected wave from another object.

Radar system overall configuration diagram A flowchart for explaining the operation of the tracking processing unit Diagram explaining the pairing area for the predicted value Diagram explaining temporary pairing between predicted value and observation point Diagram explaining overlapping pairing of observation points Diagram explaining smooth value and predicted value of next cycle The figure which shows the warning range of the own vehicle The figure explaining the measurement conditions of the tracking process (a), the figure explaining the observed value (b), and the figure explaining the tracking process result (c)

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing an example of the entire configuration of a radar system.

  In FIG. 1, the radar system 1 includes a radar ECU (Electronic Control Unit) 10 as a “radar device”, a radar 20, an engine ECU 30, a brake ECU 40, a notification device 50, and a navigation device 60. The radar ECU 10 includes a detection unit 101, a tracking processing unit 102, and an alarm processing unit 103. The radar 20 includes an antenna 201, a transceiver 202, and a signal processing unit 203 as hardware.

  As a method of the radar 20, for example, an FMCW (Frequency-Modulated Continuous Waves) radar using a frequency modulation method or a pulse radar method using a pulse modulation method can be used. For the purpose of detecting an object such as another vehicle or a bicycle in a vehicle-mounted radar system, it is possible to use, for example, (Ultra Wide Band) having a high detection capability at a distance of about 10 m from the own vehicle. it can. This makes it easy to detect an object having a small projected area when viewed from the front, such as a bicycle, and a small reflected wave intensity. The radar 20 according to the present embodiment will be described as being mounted on a moving body such as a vehicle as an example.

  The antenna 201 includes a transmission antenna that radiates a predetermined electromagnetic wave as a transmission wave, and a reception antenna that receives a reflected wave that is reflected from the object by the electromagnetic wave radiated from the transmission antenna. In FIG. 1, the antenna 201 is illustrated with the transmission antenna and the reception antenna omitted, but for example, the reception antenna may be an array antenna in which a plurality of antenna elements are arranged in an array. The array antenna is arranged side by side in the horizontal direction of the vehicle, for example, and makes it easy to specify the position in the horizontal direction of the object that reflected the reflected wave. In this embodiment, the antenna 201 is installed so as to detect a portion that becomes a blind spot of a rearview mirror or a side mirror, which will be described later with reference to FIG. For example, by installing them on the left and right of the rear part of the host vehicle, it is possible to cover the rear blind area.

  The transceiver 202 generates a reference signal serving as a reference for the transmission wave radiated from the antenna 201 and performs predetermined modulation. For example, in the case of a monopulse radar, the transmitter / receiver 202 generates monopulses with a predetermined interval (cycle) as a reference signal. The generated monopulse is modulated / demodulated by, for example, UWB modulation.

  The signal processing unit 203 mixes and amplifies the transmission wave generated by the transceiver and the reflected wave received by the antenna 201, and transmits the signal to the radar ECU 10.

  The detection unit 101, the tracking processing unit 102, and the alarm processing unit 103 of the radar ECU 10 are functional modules that realize respective functions, and any one of the functional modules can be implemented by a hardware circuit. The radar ECU 10 is a computer system including a CPU and a memory, and any one of the detection unit 101, the tracking processing unit 102, and the alarm processing unit 103 is implemented by software, and software stored in the memory is stored. It can also be implemented by executing the wear by the CPU.

  The detection unit 101 calculates the position (distance, direction) and relative speed of the object. The detection unit 101 suppresses noise by applying, for example, a low-pass filter to the signal transmitted from the radar transmission / reception unit 20. The detection unit 101 calculates the position (distance, direction) of the object from the time lag between the transmitted wave and the reflected wave received by the plurality of antennas. The detection unit 101 further performs, for example, a fast Fourier transform (FFT) process on the transmitted signal. The Fourier transform is a person who converts a temporal distribution of measured values as a frequency distribution. When the relative speed between the host vehicle and the object is positive (when the object is approaching the host vehicle), the frequency of the reflected wave is higher than the frequency of the radar transmission wave due to the Doppler effect. Conversely, when the relative speed is negative (when the object moves away from the host vehicle), the frequency of the reflected wave is lower than the frequency of the transmitted wave of the radar. Therefore, the relative speed between the subject vehicle and the object can be obtained by calculating the frequency shift between the transmitted wave and the reflected wave by fast Fourier transform.

  The tracking processing unit 102 removes unnecessary points from the detected position of the target object, for example, and calculates the current position from the target position predicted by the radar observation value one cycle before and the currently calculated target position. A filter process for estimating the position of the target object and estimating the position of the target object after one cycle is performed. Details of the operation of the tracking processing unit 102 will be described later with reference to FIG.

  The alarm processing unit 103 is provided to the driver when there is another vehicle or the like in the blind spot of the driver's rearview mirror or side mirror shown in FIG. 7 or when another vehicle is expected to enter. The warning is notified via the notification device 50 or the navigation device 60. FIG. 7 is a diagram illustrating an example of a warning range of the own vehicle.

  In FIG. 7, a radar 20L is disposed on the left rear portion of the host vehicle, and a radar 20R is disposed on the right rear portion. The radar 20L and the radar 20R observe the approach of another vehicle or the like to the left alarm range or the right alarm range surrounded by a dotted line that is a blind spot of the rearview mirror or side mirror, and the alarm processing unit 103 notifies the driver of the other vehicle or the like. The driver is warned of entering. FIG. 7 shows a state in which the bicycle overtakes the host vehicle from the left rear. In this embodiment, the progress of the approaching bicycle is predicted in advance by a tracking process described later. The alarm processing unit 103 gives an advance warning before the bicycle enters the left warning range via the notification device 50 or the navigation device 60, and further issues an entry warning when the bicycle actually enters the left warning range. Also good.

  The notification device 50 notifies the driver of a warning visually or by hearing. The notification device 50 is, for example, a display device or a sound device arranged on an instrument panel (not shown). In addition, the navigation device 60 provides the driver with route information by display or voice, and notifies an approach of another vehicle or the like to the warning range.

  The engine ECU 30 controls the engine. The engine ECU 30 communicates with the radar ECU 10 and transmits speed information of the host vehicle to the radar ECU 10. Further, the steering angle information of the host vehicle, the operation information of the direction indicator (blinker), and the like may be transmitted. The brake ECU 40 controls the brake. The radar ECU 10 gives an attention to the engine ECU 30 or the brake ECU 40 via, for example, a steering wheel or a brake pedal in order to warn of contact with another vehicle when another vehicle enters the alarm range. Also good.

  Next, details of the operation of the tracking processing unit 102 will be described with reference to FIG. 2 will be described with reference to FIGS. 3 to 6. FIG. FIG. 2 is a flowchart for explaining an example of the operation of the tracking processing unit 102.

  In FIG. 2, in the tracking process, first, a pairing candidate extraction process of an observation point and a predicted value is performed (S11). Here, the “observation point” indicates “observation position information” detected by the radar. For example, a plurality of observation points may be detected from one object. Further, the “predicted value” is “estimated position information” obtained by estimating the movement destination of the object, and one predicted value is predicted for one object. Details of the process of extracting observation point and predicted value pairing candidates will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of a pairing area of observation points with respect to a predicted value. Pairing is a process of pairing an observation point observed at time t with a predicted value predicted at time t−1, which is a time point one cycle before the radar 20, for one object.

  In FIG. 3, points A, B, and C are observation points at which the radar 20 observes a reflected wave from the bicycle at time t. Point P1 is the predicted value of the bicycle estimated at time t-1. Points Q1 and R1 are the predicted values of the guardrail poles estimated at time t-1. It is assumed that the bicycle is traveling upward in the figure from the left rear of the host vehicle at a speed faster than the host vehicle. Moreover, the pole of the guardrail is a roadside fixed object with zero speed.

  In this embodiment, the time t-1 is one cycle before the observation cycle of the radar 20 with respect to the time t. However, for example, when the radar observation cycle is short, it is a few cycles before the observation cycle. You can also.

  Next, the pairing area indicated by the dotted line in FIG. 3 is set around the point P1. The pairing area is, for example, a concentric circle around the point P1. Further, the pairing area can be expanded from a concentric circle centered on P1 to a range close to the radar 20. Further, if the bicycle observed at time t-1 is accelerating (when the relative speed with the host vehicle is increasing), the pairing area can be expanded in the traveling direction of the bicycle, that is, upward in the figure. Conversely, if the bicycle is decelerating, the pairing area can be expanded downward in the figure. Since observation points A to C are in the pairing area of point P1, they are extracted as pairing candidates.

  Returning to FIG. 2, it is determined whether or not there are a plurality of observation points at time t in the pairing area (S12). When there is one observation point in the pairing area (N in S12), the observation point is provisionally paired with the predicted value (S16).

  Here, when there is no observation point in the pairing area (N in S12), if an observation point to be paired is not detected during a predetermined cycle, tracking at the prediction point is failed and the next cycle The predicted value at may not be calculated. Further, an unobserved section may be supplemented from the tracking history. Further, the pairing area may be enlarged.

  When there are a plurality of observation points in the pairing area (Y in S12), it is determined whether or not a predetermined calculation formula is larger than a threshold value (S13). In FIG. 3, since there are three observation points from point A to point C in the pairing area of point P1, step S12 is Y. Here, the calculation formula is exemplified below.

α · Own vehicle speed + β · Predicted value relative speed> Threshold value (1)
In equation (1), α and β are predetermined coefficients for weighting. Since the predicted value relative speed in equation (1) is (absolute velocity of predicted value−vehicle speed), if α = β, equation (1) is the absolute value of the predicted value, that is, the speed of the bicycle Is equivalent to determining whether or not is greater than a threshold.

  In the equation (1), when the own vehicle speed is low, both the stationary object and the moving object are observed at the observation point of the radar 20, whereas when the own vehicle speed is high, the comparison is made when the own vehicle speed is low. Then, since the observation point of the stationary object is immediately outside the measurement range of the radar, the measurement value is observed at a position that varies according to the presence of the roadside fixed object. On the other hand, an object having a close relative speed is detected relatively stably. Therefore, when the expression (1) is larger than the threshold value (Y in S13), that is, when the own vehicle speed is high or the predicted point relative speed of the target object is low, for example, as the target of temporary pairing, An observation point having the largest relative speed (the largest velocity vector in the direction of the host vehicle) is set as a target for temporary pairing (S14, S16). On the other hand, when the expression (1) is equal to or less than the threshold value (N in S13), that is, for example, when the own vehicle speed is low or the predicted point relative speed of the target object is low, the nearest observation point from the traveling line of the own vehicle is selected. The target of temporary pairing is set (S15, S16). Here, the details of the method of selecting the provisional pairing target in step S15 will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of provisional pairing between a predicted value and an observation point when step S13 is N.

  In FIG. 4, when the own lane is a trajectory at the center of the radar 20, the distance between the own lane and the points A to C is xA, xB, and xC, respectively. Here, the closest point to the own lane is a point C having a distance xC. Therefore, the point P1 is provisionally paired with the point C.

  In addition, the process of step S11 to S16 of FIG. 2 is performed with respect to all the predicted values. That is, temporary pairing of observation points is performed for all predicted values.

  Next, it is determined whether there is an observation point paired with a plurality of predicted values (S17). When the observation point paired with the predicted value does not overlap (N in S17), the temporary paired state is used as a pair for pairing (S21). On the other hand, when there are observation points paired with a plurality of predicted values (Y in S17), the overlapping pairing is eliminated by the method determined by the equation (1) described in step S13.

  When the formula (1) is larger than the threshold value (Y in S18), the predicted value having the largest relative speed among the predicted values (the largest speed vector in the direction of the host vehicle) is set as the target of pairing. Pairing with the predicted value is canceled (S19, S21). On the other hand, when the formula (1) is equal to or less than the threshold (N in S18), pairing with other predicted values is canceled with the predicted value closest to the traveling line of the host vehicle as the target of pairing (S20, S21). ). Here, the elimination of overlapping pairing will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of overlapping pairing of observation points.

  In FIG. 5, point C, which is an observation point, is paired from both the predicted value point P1 of the bicycle and the predicted value point R1 of the guardrail pole. Since the point P1 is a traveling vehicle, the relative speed is higher than the point R1 of the roadside fixed object. Further, since the distance xP between the own lane and the point P1 is shorter than the distance xR with the point R1, the point P1 becomes a pairing target regardless of the magnitude relationship with the threshold value in step S18, and the pairing with the point R1 Is resolved.

  Returning to FIG. 2, the predicted value and the value of the observation point for which the pair is determined are added to the tracking filter (S22). Here, the filter is, for example, a Kalman filter that calculates a smoothed value that is an output obtained by correcting the current observation point from the past prediction value and the current observation point, and a future prediction value. The Kalman filter is a filter that corrects an observation point including a measurement error, such as a radar reflected wave, from a predicted value in the past, and further predicts a future value to converge the error. In the present embodiment, the object of pairing is changed by the determination based on the speed described in Expression (1), and the error of the observation point due to interference of reflected waves is reduced. By the tracking filter process (S22), a predicted value is calculated (S23), and a smooth value (S24) is further calculated. The predicted value and smooth value calculated in steps S23 and S24 will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of the smooth value and the predicted value of the next cycle.

  In FIG. 6, a point L is a smooth value at time t calculated in step S22 by substituting the point C paired with the point P1 into a Kalman filter. Point P2 is a predicted value of the bicycle at time t + 1 estimated at time t calculated from points P1 and C.

  In the present embodiment, the predicted value and the observation point to be substituted into the Kalman filter are appropriately selected for each measurement cycle of the radar 20, so that the observation point is an object that has a weak reflected wave intensity, such as a bicycle, and is observed every cycle. Even when a measurement error is included in the object, tracking of the object can be performed accurately. In addition, even when there is an object having a high reflected wave intensity such as a roadside fixed object, tracking can be continued with priority on an object such as a bicycle.

  The predicted value calculated in step S23 is processed as a pairing target for the next cycle (S11). This completes the tracking processing operation.

  When the tracking process predicts that the predicted value or smooth value of the target bicycle falls within the warning range described in FIG. 7, the warning device 50 in FIG. 1 notifies the driver of the warning. The

  Next, the tracking result of the object in the present embodiment will be described with reference to FIG. FIG. 8A is a diagram for explaining an example of measurement conditions for tracking processing in the present embodiment, FIG. 8B is a diagram for explaining an example of observation values, and FIG. 8C is a diagram for explaining an example of tracking processing results. is there.

  In FIG. 8A, the host vehicle is stopped so that the distance from the left end of the host vehicle to the guard rail is 5 m from the guard rail parallel to the traveling direction in the x direction shown in the drawing. Pole is installed at equal intervals on the guardrail. A radar 20 is attached to the left rear of the host vehicle. With this arrangement, the bicycle goes straight at a constant speed at a distance of 3 m from the guardrail. FIG. 8B is a plot of observation values (points) of the radar 20 under these measurement conditions.

  In FIG. 8B, the x-axis (time t) is the time in one processing cycle, and the y-axis (tY) is the distance from the host vehicle at time t in the y-direction in FIG. 8A. Here, the solid line in the graph is a true value indicating the actual traveling locus of the bicycle. Further, a range of ± 1 m indicated by a broken line centered on the true value is set as a correct observation point, and other detected observation points are deleted from the graph. In FIG. 8 (b), it can be seen that the observation points are interrupted in the predetermined range shown in the y direction. This is because the reflected wave from the radar reflected from the bicycle interferes with the strong reflected wave reflected from the guardrail. FIG. 8C is a plot of the result of the tracking process described in FIG.

  In FIG. 8 (c), the range that was interrupted in FIG. 8 (b) is partially tracked although it is partially interrupted. Since the measurement described in FIG. 8 is in a state where the host vehicle is stopped, the reflected wave from the guard rail is fixed and the reflection from the girl rail is strongly influenced. In the present embodiment, it is shown that the tracking performance of a bicycle or the like can be improved even when there is a fixed object such as a guardrail that generates a strong reflected wave.

  As mentioned above, although the form for implementing this invention was explained in full detail, this invention is not limited to such specific embodiment, In the range of the summary of this invention described in the claim, Various modifications and changes are possible.

  For example, instead of setting the pairing target to be one-to-one between the predicted value and the observation point, the predicted value and the average value of a plurality of observation points can be paired.

  Further, the intensity of the reflected wave is measured at the observation point, and weighting based on the intensity of the reflected wave can be applied when selecting the observation point.

1 Radar device 10 Radar ECU
DESCRIPTION OF SYMBOLS 101 Detection part 102 Tracking processing part 103 Alarm processing part 20 Radar transmission / reception part 201 Antenna 202 Transceiver 203 Signal processing part 30 Engine ECU
40 Brake ECU
50 Notification device 60 Navigation device

Claims (7)

A detection unit that receives a signal at a first time point from a radar that observes a surrounding object in a first cycle, and detects observation position information of the object from the received signal;
Estimating estimated position information of the object at a second time point after the first cycle from the first time point from the observation position information detected by the detection unit, and a pair of the estimated position information and the estimated position information A tracking unit that pairs the observation position information at the second time point within the ring range and estimates the estimated position information of the object at the third time point after the first cycle from the second time point; A radar apparatus comprising:   When the observation position information at a plurality of second time points is detected within the pairing range, the tracking processing unit observes at the second time point at which the distance from the own lane indicating the traveling direction of the own vehicle is the closest. The radar apparatus according to claim 1, wherein first pairing that prioritizes pairing with position information is performed. The detector further detects a relative speed between the host vehicle and the object,
The tracking processing unit, when the observation position information at a plurality of second time points is detected within the pairing range, the observation position at the second time point at which the speed approaching the host vehicle is highest from the relative speed. The radar apparatus according to claim 1 or 2, wherein second pairing is performed in which pairing with information is prioritized.   4. The radar device according to claim 2, wherein the tracking processing unit switches between the first pairing and the second pairing according to a speed of the host vehicle and the relative speed. 5.   The tracking processing unit performs pairing according to the speed of the host vehicle and the relative speed when there are a plurality of estimated position information at the second time point paired with the observation position information at the second time point. The radar apparatus according to claim 1, wherein estimated position information at the second time point is selected.   2. The tracking processing unit outputs estimated position information at the third time point and a smooth value at the second time point from the paired estimated position information and observed position information at the second time point. The radar apparatus as described in any one of thru | or 5. A receiving process for receiving a signal at a first time point from a radar that observes surrounding objects in a first cycle;
Detection processing for detecting observation position information of the object from the signal received in the reception processing;
A first estimation process for estimating estimated position information of the object at a second time point after the first cycle from the first time point, based on the observation position information detected by the detection process;
Pairing processing for pairing the estimated position information and the observed position information at the second time point within the pairing range of the estimated position information;
A radar apparatus control method in which a computer executes a second estimation process for estimating estimated position information of the object at a third time point after the first cycle from the second time point.

Images