JP2001319299A - Road curvature estimating device for vehicle and preceding vehicle selecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique by which the curvature of a road in a front part is properly estimated even in a road which has a curve in the front part though a vehicle is traveling along a straight part. SOLUTION: A curvature radius is calculated based on a changer rate when the change with time of an object horizontal position is linearly approximated. Since a triangle OAB consisting of a curvature center O, its own vehicle A and an object B is an isosceles triangle, θ+2ψ=π, and θ=2ϕ from ϕ+ψ=(1/2) π. Since Rsinθ is Dcosϕ and Vx being the X-direction component of the relative velociaty V of an object to be recognized is Vsinθ, R=Dcosϕ×V/Vx is obtained. R=Dvsens/Vx is obtained since Vsens being the relative velocity of the object B is equal to Vcos(θ-ϕ) and to Vcosϕ. Vx being the X-direction component of the relative velocity V is dX/dt. When it is replaced with a different ΔX/Δt, the inclination (a) of a graph which indicates the locus of the object to be recognized is obtained. Therefore, the curvature radius = (a distance to the object to be recognized) × (the relative velocity)/the inclination.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for irradiating a transmission wave within a predetermined angle range in a vehicle width direction and estimating a curvature of a road ahead of the vehicle based on a reflected wave.

[0002]

2. Description of the Related Art Conventionally, a transmission wave such as a light wave or a millimeter wave is radiated over a predetermined angle in front of a vehicle, and a reflected wave is detected. An obstacle recognition device for a vehicle that recognizes an obstacle has been considered. This type of device is applied to, for example, a device that detects an obstacle such as a preceding vehicle and generates an alarm, and a device that controls a vehicle speed so as to maintain a predetermined inter-vehicle distance with the preceding vehicle, and is a control target. A vehicle that recognizes an obstacle such as a preceding vehicle has been considered.

In this vehicle obstacle detecting device, it is necessary to specify a preceding vehicle traveling in front of the own vehicle from the recognized obstacles. For this purpose, for example, a steering angle sensor, a yaw rate sensor, or the like is used. It is common. However, for example, if your car is traveling on a straight part of the road,
If the road curves ahead and the preceding vehicle is already in the curve, the preceding vehicle may be erroneously identified. In order to properly identify the preceding vehicle in such a situation, it is necessary to appropriately estimate the curvature of the road ahead of the vehicle.

[0004] Therefore, the present invention provides a technique capable of appropriately estimating the curvature of a road ahead even on a road where the vehicle is traveling in a straight line but curves ahead. An object of the present invention is to provide a technique for selecting a preceding vehicle based on an estimated road curvature.

[0005]

According to a first aspect of the present invention, there is provided a road curvature estimating apparatus which irradiates a transmission wave over a predetermined angle range in a vehicle width direction, and calculates a road curvature ahead of the vehicle based on the reflected wave. For estimation, the following processing is performed. That is, at least the distance to the object and the object lateral position are calculated, the change rate of the object lateral position is calculated from the history of the calculated object lateral position over time, and the vehicle is calculated based on the calculated change rate. It calculates the curvature of the road ahead.

Here, the "object lateral position" refers to how far the object is located in the vehicle width direction with respect to an extension line of the longitudinal axis (front direction) of the own vehicle. It is a physical quantity shown. The object lateral position of the preceding vehicle on the own lane on a straight road basically does not change. The term "basically" is used in consideration of, for example, a slight change even in the same lane. On the other hand, for example, as shown in FIG. 2A, in a case where the own vehicle is traveling on a straight line portion, but is on a curved road ahead, the preceding vehicle travels on the curved portion. In such a case, the object lateral position changes even on the same lane. That is, FIG.
In the situation of (a), since the preceding vehicle is traveling in the right lane and has a right curve, the preceding vehicle's object lateral position when both vehicles are traveling in a straight line in the same manner, When only the preceding vehicle is traveling on a curved portion, the object lateral position of the preceding vehicle is larger. Therefore, if attention is paid to such a temporal change in the lateral position of the object, it is considered that the curvature of the road ahead can be appropriately estimated.
Of course, not only a moving object but also a stationary object such as a roadside object can similarly appropriately estimate the curvature of the road ahead based on the change rate of the lateral position of the object. FIG.
(B) and (c) show graph examples of temporal changes (trajectories) of the object lateral position in the case of a left curve and in the case of a right curve.

As for the calculation of the rate of change, it is conceivable that the history of the lateral position of the object over time is approximated by a straight line, as described in claim 2. The fact that the curvature can be appropriately calculated by this method will be described below. In calculating the curvature, it is general to calculate the radius of curvature (R), and therefore, it will be described with reference to FIG. 3 that the radius of curvature R is appropriately calculated.

As shown in FIG. 3A, the curvature of a curved road when the vehicle A travels on a curved road having a radius of curvature R and recognizes a target B in the curved road is as follows. Required. First, consider △ OAB including the origin O, the host vehicle A, and the target B in FIG. The origin O is the center of curvature, the X-axis is the width direction of the host vehicle, and the Z-axis is the same direction as the front-rear direction. In this $ OAB, $ AO
Assuming that B = θ and ∠OAB = ψ, since で あ OAB is an isosceles triangle, ∠OBA = ∠OAB = ψ.
The following equation 1 holds.

Θ + 2ψ = π [Equation 1] When the angle between the tangent to the arc of the curvature radius R at the point A and the line segment AB is φ, the following equation 2 is established. φ + ψ = (1/2) π (Expression 2) From Expressions 1 and 2, the relationship of θ = 2φ is derived.

If the length of the line segment AB is D, the following equation 3 holds. R sin θ = D cos φ (Equation 3) Here, assuming that the recognized object has a relative velocity V in a tangential direction of an arc having a radius of curvature R, its X-direction component Vx is Vx = V
sinθ. Therefore, using this relationship,
When nθ is eliminated, the following Expression 4 is established.

R = Dcosφ · V / Vx (Equation 4) The relative speed Vsens of the target B sensed from the host vehicle A is expressed by the following Equation 5. Vsens = Vcos (θ−φ) = Vcosφ (Expression 5) When this is substituted into Expression 4, the following Expression 6 is obtained.

R = DVsens / Vx [Equation 6] Here, Vx which is the X-direction component of the relative speed V is dX /
dt, and when this is replaced with the difference ΔX / Δt, it becomes the slope of the graph showing the trajectory of the recognized target (see FIG. 3B). For example, a target is moved to a horizontal position X0 at time t = t0.
Recognition starts from [m], the horizontal position at the time of performing recognition for Δt [sec] is Xt [m], the distance to the target at that time is d [m], and the relative speed is v [m]. ]. If the locus of the recognized target is linearly approximated and expressed as X (t) = at + b, then Vx
= A [m / sec].

Since D and Vsens are the distance to the target and the relative speed, respectively, the radius of curvature R can be obtained as in the following equation (7). R = dv / a (Equation 7) By the way, when calculating the rate of change, at least in the first and second data are detected continuously at the initial stage of object detection. Only if
The rate of change may be calculated. This requires a plurality of data (horizontal position data) to calculate the rate of change, but it is premised that the plurality of data relate to the same object. In view of this point, if there is the first and second data at the initial stage of detection, even if the third data is not detected and the fourth data is detected, they are the same object. It is easy to determine that this is the detection result. This is because the first data exists but the second
There is no data for the first time, and even if the data for the third time is detected, it is doubtful whether the data for the first time and the data for the third time are a detection result on the same object. That is, it is highly likely that what is considered the third data is actually data of another object. Therefore, by imposing such conditions, more appropriate curvature estimation can be performed.

Although the curvature of the road can be estimated for any of a stationary object and a moving object, a stationary object is preferable.
Because, in the case of a moving object, its speed changes,
Alternatively, the lane may be changed, and the traveling position may move left and right even in the same lane. Therefore, if the vehicle is a stationary object, the estimation error due to such a change need not be considered. . Therefore, as described in claim 4, when the recognition type of the moving object or the stationary object can also be determined, if the stationary object is recognized, the change rate using the lateral position of the stationary object is calculated, When the stationary object is not recognized but the moving object is recognized, it is conceivable to calculate the rate of change using the lateral position of the moving object.

Also, the curvature calculated by the curvature calculation means is not directly used as the road curvature, but an average value for a predetermined number of times is calculated, and the average value is used as the road curvature. It is also possible to do. By taking the average value in this way, the value is smoothed, and the influence of temporary noise data can be prevented.

When the condition for calculating the rate of change is not satisfied when taking the average value in this way, the curvature using the rate of change cannot be calculated. For example, the change rate is based on the temporal change of the horizontal position, and thus there is a case where a sufficient past history of the horizontal position is not obtained to specify the temporal change. Therefore, in order to calculate the average value even in such a situation, the invention may be configured as in claim 6.
That is, there is provided a second curvature estimating means for detecting a turning state of the own vehicle by a turning detecting means such as a steering sensor or a yaw rate sensor, and estimating a curvature of the road based on the turning state and the own vehicle speed. When the condition that the change rate calculating means can calculate the change rate is not satisfied, the average value calculating means adopts the curvature estimated by the second curvature estimating means and performs the predetermined number of times including the curvature. Is calculated.

On the other hand, when a second curvature estimating means for estimating the curvature of the road based on the turning state of the own vehicle and the own vehicle speed is provided, the second curvature estimating means is provided.
The final road curvature may be estimated based on the curvature estimated by the curvature estimating means and the curvature estimated based on the change rate of the lateral position of the object. Since the technique of "estimation" is used, this is effective from the viewpoint of avoiding erroneous determination as much as possible.

Further, the present invention is provided with the vehicle road curvature estimating apparatus described above, and based on the road curvature obtained from the apparatus and the relative position of the object, the object is positioned on the same lane as the own vehicle. , And a preceding vehicle selecting device that selects a preceding vehicle based on the probability. As a method of selecting a preceding vehicle based on the own lane probability, for example, Japanese Unexamined Patent Application Publication No.
No. 9099 can be considered.

On the other hand, the preceding vehicle may be selected based on the road curvature for each object obtained by the above-described vehicle road curvature estimating apparatus instead of the above-described own lane probability. That is, according to the ninth aspect, the road curvature estimating device for a vehicle obtains a road curvature based on a stationary object which is assumed to exist at both ends of the road. If the road curvature is obtained, the possibility that the moving object is in the same lane as the own vehicle is determined based on the magnitude relationship between the road curvature and the road curvature obtained based on the moving object. I do.
Then, the preceding vehicle is selected based on the determination result.

The concept of this technical idea will be described with reference to FIG. Here, a two-lane road is assumed as shown in FIG. There is a stop 1 and a stop 2 at the left and right road edges, respectively, in a portion where the road curves right, a moving object 1 exists in the right lane, and the vehicle moves in the same lane as the own vehicle traveling in the left lane. Consider a situation in which the object 2 is present and the moving objects 1 and 2 are both traveling on a curved portion after T time.
FIG. 8B shows each of the recognition targets (moving objects 1 and 2 and stationary objects 1 and 2).
It is a graph which shows the locus | trajectory of the horizontal position of 2). In the case of a moving object, the rate of change in the lateral position differs depending on the lane in which the vehicle is traveling.
This is because the curvature (radius) differs depending on the traveling lane. In addition, the rate of change of the lateral position of the stationary object differs depending on the existing position. Similarly, the curvature (radius) is different.

Therefore, as shown in FIG. 8A, the radius of curvature of the moving object 2 after T time is R1, the radius of curvature of the stopped object 1 is R2, the radius of curvature of the stopped object 2 is R3, and the time after T time is elapsed. Assuming that the radius of curvature of the moving object 1 is R4, their magnitude relation is R3
<R4 <R1 <R2. By utilizing such a magnitude relationship, it can be determined whether or not the moving object is on the same lane, and the preceding vehicle can be selected.

It should be noted that whether or not the road curvature is based on a stationary object which is assumed to be present at each end of the road is determined as follows.
The determination may be made based on whether or not the difference between the road curvature radii is equal to or larger than a predetermined value (claim 10). For example, if the curvature radius difference is equal to or larger than the assumed road width, it is considered that the stationary object exists at both ends of the road. If the number of lanes of the road on which the vehicle is traveling is known, a predetermined value may be determined based on the road width according to the number of lanes.

Further, as a method of determining the same lane, for example, the following method can be considered. That is,
Based on the object lateral positions of the stationary objects that are assumed to be present at both ends of the road, it is determined whether the vehicle is on the right or left side of the road. And, as a result of the determination,
The determination is made on the basis of the magnitude relationship between the curvature of the road based on the moving object to be determined whether the lanes are the same or not and the curvature of the road based on the stationary objects assumed to be present at both ends of the road.

The curvature estimating means, the average value calculating means, and the second
The function of realizing the curvature estimating means, the own lane probability calculating means, the preceding vehicle selecting means, and the same lane determining means in the computer system can be provided as, for example, a program started on the computer system side. In the case of such a program, for example, it is recorded on a computer-readable recording medium such as a floppy (registered trademark) disk, a magneto-optical disk, a CD-ROM, and a hard disk, and is loaded into a computer system as needed and activated. Can be used. Alternatively, the program may be recorded in a ROM or a backup RAM as a computer-readable recording medium, and the ROM or the backup RAM may be incorporated in a computer system and used.

[0025]

Next, a vehicle control device 1 to which the present invention is applied will be described with reference to the drawings. The vehicle control device 1 is mounted on an automobile and outputs a warning when an obstacle exists in a region to be warned in a predetermined situation, or controls a vehicle speed in accordance with a preceding vehicle (preceding vehicle). It is.

FIG. 1 is a system block diagram of the system.
The vehicle control device 1 mainly includes a computer 3. The computer 3 mainly includes a microcomputer and includes an input / output interface (I / O) and various drive circuits and detection circuits. Since these hardware configurations are general, detailed description will be omitted.

The computer 3 includes a radar device 5, a vehicle speed sensor 7, a brake switch 9, and a throttle opening sensor 1.
1, predetermined detection data is input. Further, the computer 3 outputs a predetermined drive signal to the alarm sound generator 13, the distance indicator 15, the sensor abnormality indicator 17, the brake driver 19, the throttle driver 21, and the automatic transmission controller 23.

Further, the computer 3 detects an alarm volume setting device 24 for setting an alarm volume, an alarm sensitivity setting device 25 for setting a sensitivity in an alarm determination process described later, a cruise control switch 26, and an operation amount of a steering wheel (not shown). A steering sensor 27 and a yaw rate sensor 28 are provided. Further, the computer 3 includes a power switch 29, and starts a predetermined process when the power switch 29 is turned on.

Here, the radar device 5 will be described. The radar device 5 transmits a radar wave modulated to a predetermined frequency according to the modulation signal, and receives a radar wave reflected by an obstacle, a transmission / reception unit 5a, and supplies a modulation signal to the transmission / reception unit 5a. The signal processing unit 5b executes a process for detecting an obstacle based on the intermediate frequency beat signal output from the transmission / reception unit 5a. In this embodiment, in order to detect an obstacle in front of the vehicle by the device 5, the transmitting / receiving unit 5a is mounted on the front of the vehicle, and the signal processing unit 5b is mounted at a predetermined position in or near the vehicle. Have been.

The transmitting antenna included in the transmitting / receiving section 5a is such that the transmitting beam can cover the entire detection range in which the radar device 5 can detect an obstacle. The receiving antenna that receives radar waves is one that forms a highly directional receiving beam, and is configured to be able to rotate the irradiating direction of the receiving beam in any direction along the horizontal plane. ing. Then, the transmitting / receiving unit 5a performs scanning within the above-described detection range by sequentially switching the irradiation direction of the receiving beam in units of a predetermined step angle according to the control signal from the signal processing unit 5b.

On the other hand, the signal processing section 5b includes a triangular wave generator for generating a modulation signal, an A / D converter for converting a beat signal from the transmission / reception section 5a into digital data,
OM, RAM, triangular wave generator, A / D
A well-known microcomputer (microcomputer) that executes a data collection process of collecting data by operating a converter and the like, and an obstacle detection process of detecting an obstacle based on digital data collected by the data collection process. ), And an arithmetic processing unit that executes an operation of fast Fourier transform (FFT) based on a command from the microcomputer. Then, the receiving antenna of the transmission / reception unit 5a is moved to an initial position (for example, a position where the receiving beam comes to one end in the detection range), and subsequently, a predetermined value is determined according to the rising gradient of the triangular waveform of the modulation signal. The transmission signal is modulated so that the frequency increases (upward modulation section) at a rate of and the frequency decreases (downward modulation section) in accordance with the subsequent downward gradient. That is, the radar device 5 of the present embodiment
Operates as a so-called FMCW radar. When a radar wave corresponding to the frequency-modulated transmission signal is transmitted and a radar wave reflected on an obstacle is received, the transmission / reception unit 5a outputs a reception antenna. A beat signal is generated by mixing the reception signal and the transmission signal output from the control unit. Note that the received signal is delayed with respect to the transmitted signal by the time that the radar wave travels back and forth to the obstacle, and if there is a relative speed between the radar signal and the obstacle, the received signal is subjected to a Doppler shift according to the relative speed. . For this reason, the beat signal includes the delay component and the Doppler component.

The beat signal is A / D converted, and the digital data obtained as a result is sequentially stored in the RAM in the signal processing unit 5b. The A / D converter performs A / D conversion a predetermined number of times while the modulated signal is being output. That is, the first half of the data is the uplink modulation data corresponding to the uplink modulation section of the transmission signal, and the second half of the data is the downlink modulation data corresponding to the downlink modulation section of the transmission signal. When the processing for one cycle of the modulation signal is completed in this way, the irradiation direction of the receiving beam of the receiving antenna is moved by one step, and the above processing is repeatedly executed. Then, when scanning of the reception beam is completed over the entire detection range, an obstacle detection process is started,
Thereafter, the present process is repeatedly executed in parallel with the obstacle detection process.

In the obstacle detection processing, digital data written in the RAM is sequentially read out, and a fast Fourier transform (FFT) operation is performed for each reception beam and each modulation unit (up / down modulation unit). By doing so, the frequency analysis of the beat signal is performed. Then, the peak of the frequency spectrum is detected, the frequency is specified as a beat frequency (fu in the up-modulation unit, fd in the down-modulation unit), and the detected frequency component is compared with the amplitude of each received beam. From the signal intensity distribution, a peak receiving beam is specified. Then, the irradiation direction of the specified reception beam is set as angle data representing the arrival direction of the radar wave, that is, the direction in which the obstacle exists. However, the angle data is defined such that the clockwise direction is represented by a positive value and the counterclockwise direction is represented by a negative value, with the traveling direction in front of the vehicle being 0 °. And
When there are a plurality of peak frequency components, that is, when a plurality of obstacles have been detected, the up-modulation unit and the down-modulation unit perform a pairing process of correctly combining frequency components based on the same obstacle. This pairing process is performed by, for example, combining the components in the order of lower frequency or combining components having substantially the same signal strength.
In this pairing process, the identity is also determined by comparing the information with information on an obstacle detected when the process was last activated. Then, distance data to the obstacle and relative speed data are calculated from the beat frequencies fu and fd of the rising part and the falling part.

The description of the radar device 5 has been described above, and returns to the description of the computer 3. The computer 3 performs an alarm determination process for alarming when an obstacle exists in a predetermined alarm area for a predetermined time based on information from the radar device 5 or the like. As the obstacle, a preceding vehicle traveling in front of the own vehicle, a preceding vehicle that is stopped, or an object (a guardrail or a support object) on the roadside corresponds to the obstacle. The computer 3 also outputs so-called drive signals to the brake driver 19, the throttle driver 21, and the automatic transmission controller 23, thereby simultaneously controlling the vehicle speed according to the situation of the preceding vehicle, that is, so-called inter-vehicle control. ing.

Next, the internal configuration of the computer 3 will be described as a control block. The distance D output from the signal processing unit 5b of the radar device 5 and the irradiation angle θ of the reception beam
And the data of the relative speed V are sent to a coordinate conversion block 41 between polar coordinates and rectangular coordinates. Here, the laser radar center is set to the origin (0, 0), and is converted into XZ orthogonal coordinates having the vehicle width direction as the X axis and the vehicle front direction as the Z axis. Is output. It should be noted that the same is output in the state of the data of the distance D and the relative speed V.

In the object recognition block 43, the center position of the object (X,
Z). The value of X becomes a value indicating the “object lateral position”. In the object recognition block 43, the vehicle speed (own vehicle speed) output from the vehicle speed calculation block 47 based on the detection value of the vehicle speed sensor 7 and the relative speed (Vx, V
z), a recognition type of whether the object is a stationary object or a moving object is obtained, and an object that affects the traveling of the own vehicle is selected based on the recognition type and the center position of the object. It is displayed by the distance indicator 15.

The sensor abnormality detection block 44 detects whether the data obtained in the object recognition block 43 has a value in an abnormal range. If the data is in an abnormal range, the sensor abnormality display 17 displays the value. Is displayed. Also,
A steering angle is obtained in a steering angle calculation block 49 based on a signal from the steering sensor 27, and a yaw rate calculation block 5 based on a signal from the yaw rate sensor 28.
At 1, the yaw rate is calculated. The curvature radius calculation block 63 calculates a curvature radius based on the vehicle speed from the vehicle speed calculation block 47, the steering angle from the steering angle calculation block 49, and the yaw rate from the yaw rate calculation block 51.

On the other hand, in the road curvature estimating block 45, the measured data converted into the rectangular coordinates and the object recognition block 4 are used.
The road curvature is estimated based on the data obtained in step 3 and the radius of curvature based on the steering angle and the like obtained in the radius of curvature calculation block 63. Details of the road curvature estimation processing will be described later. The data obtained by the road curvature estimation block 45 is output to the preceding vehicle determination block 53.

In the preceding vehicle determination block 53, the estimated radius of curvature R, the type of recognition obtained in the object recognition block 43, the center position coordinates (X, Z), and the relative speed (V
(x, Vz), a preceding vehicle is selected, and a distance Z and a relative speed Vz to the preceding vehicle are determined.

Then, the inter-vehicle control unit and the alarm determination unit block 55 determine the distance Z to the preceding vehicle, the relative speed Vz, the own vehicle speed Vn, the preceding vehicle acceleration, the object center position, the object width, the recognition type, and the cruise control switch 26. Based on the setting state of the brake switch 9, the depression state of the brake switch 9, the opening from the throttle opening sensor 11, and the sensitivity set value by the alarm sensitivity setting device 25, it is determined whether or not to give an alarm if it is an alarm judgment, For example, the content of the vehicle speed control is determined. The result is output to the alarm sound generator 13 if an alarm is required. If the cruise is determined, a control signal is output to the automatic transmission controller 23, the brake driver 19, and the throttle driver 21 to perform necessary control. When these controls are executed, a necessary display signal is output to the distance display 15 to notify the driver of the situation.

Next, the operation for estimating the road curvature, which is performed in the vehicle control device 1 configured as described above, will be described with reference to the flowchart of FIG. FIG.
In the first step S10, radar signal processing is performed. This processing is performed by the radar device 5. As described above, the read of the beat signal, the frequency analysis of the beat signal by the FFT operation, the extraction of the target peak by the peak search, the pairing of the extracted peak, and the like are performed. Do. Then, in S20, a target recognition process is executed.
In this process, as described above, the distance D to the obstacle and the relative speed V are calculated from the beat frequencies fu and fd of the ascending and descending portions, and the lateral position of the object is calculated from the distance D and the irradiation angle of the reception beam. X is calculated. The recognition result in S20 is stored in the memory prepared in the object recognition block 43 (S30).

At S40, the vehicle speed calculated by the vehicle speed calculation block 47 is read. At S50, the road curvature is calculated based on the steering angle obtained from the steering sensor 27 (or the yaw rate obtained from the yaw rate sensor 28). Here, since it is basically calculated using the steering angle, it is referred to as an estimated R based on the steering angle. We ask as follows. Estimated R based on steering angle = f (own vehicle speed) ÷ steering angle Note that the function using the own vehicle speed as a variable for obtaining the estimated R (curve radius) from the steering angle is a function determined from the vehicle motion characteristics. , Which is generally known to those skilled in the art, and a detailed description thereof will be omitted.

The following S70 and S80 are processes for estimating the road curvature from the recognized target.
It is determined whether the recognition result in S20 stored in the object recognition block 43 in S30 satisfies a certain condition. In the present embodiment, the fixed condition is a condition that “at least the first and second data are continuously detected at the initial stage of object detection”. This is a condition from the following intentions. That is, although the road curvature is estimated based on the change rate of the lateral position of the target in S80, a plurality of data (lateral position data) are required to calculate the change rate, and the plurality of data are It is assumed that they are related to the same object. In view of this point, if there is the first and second data at the initial stage of detection, even if the third data is not detected and the fourth data is detected, they are the same object. It is easy to determine that this is the detection result. This is because even if the first data is present but the second data is not present and the third data is detected, the first and third data are finally detected on the same object. It is questionable. That is, it is highly likely that what is considered the third data is actually data of another object. Therefore, imposing such conditions enables more appropriate curvature estimation.

When such a condition is satisfied (S70: Y
In ES), the flow shifts to S80 to perform a road curvature estimation process based on the recognized target, and then shifts to S90. On the other hand, when the condition is not satisfied (S70: NO), the process proceeds to S90 without executing the process of S80.

The processing in S80 is performed by the road curvature estimation block 45.
The details of the process in S80 will be described with reference to FIGS. In the first step S310 in FIG. 5, the relative speed V, the lateral position X, and the distance D are extracted as the data of the recognized target. Then, when at least one stop exists in the recognition target (S320:
YES), the process proceeds to S330, and the number of stopped objects is M, and S
The processing of 340 to S390 is repeated M times. Specifically, the number of consecutive recognitions exceeds n (S340: Y
ES), and if the recognition start distance exceeds d (S350: YES), the flow shifts to S360; otherwise, the flow shifts to S380.

In S360, based on the history of the horizontal position X _M (t) data, a straight line approximation is performed by using a linear equation X _M (t) = at + b, and a slope a is obtained. Then, in S370, using the distance D and the relative speed V of the target history data and the inclination a of the straight line obtained in S360, the curvature radius R (M) of the road is used.
Is determined by R (M) = DV / a. The reason why the radius of curvature is obtained by this equation has been described above, and will not be repeated here. On the other hand, in S370, the radius of curvature estimated to the R (M) by the steering angle or the like obtained by the steering sensor 27 (or the yaw rate may be used as described above), that is, the radius of curvature calculated by the radius of curvature calculator 63 (see FIG. 1). Substitute the radius of curvature.

After the processing of S370 or S380, the number M of stopped objects is decremented (M = M-1) (S390), and M
Is larger than 0 (S400: YES), S340
When M = 0 (S400: NO), the process returns to S37.
The average value of the M R (M) obtained in the processing of 0 or S380 is obtained, and the obtained average value is set as the radius of curvature based on the target.

As described above, if there is a stationary object (S320: YES), the radius of curvature based on the target is obtained by giving priority to the stationary object, but if there is no stationary object (S32).
0: NO), and proceeds to S420 to perform processing for a moving object. The contents of the process in S420 will be described with reference to FIG.

In the first step S510 in FIG. 6, the number of targets for which the relative speed V is positive is set to M. In the subsequent S350,
It is determined whether M> N, and if M> N (S5
20: YES), the process of S530 to S590 is repeated M times. More specifically, if the number of continuous recognitions exceeds n (S530: YES) and the recognition start distance exceeds d (S540: YES), the process proceeds to S550, otherwise. Shifts to S580. S55
At 0, the linear formula X _M (t) is based on the history of the horizontal position X _M (t) data.
_A linear approximation is performed using _M (t) = at + b, and a gradient a is obtained. Then, in S560, using the distance D and the relative speed V of the target history data and the gradient a of the straight line obtained in S550, the curvature radius R (M) of the road is calculated as R (M) = DV /
Obtain with a. In S580, the radius of curvature estimated from the steering angle or the like obtained by the steering sensor 27 is substituted for R (M).

After the processing of S560 or S580, the number M of stopped objects is decremented (M = M-1) (S690), and M
Is larger than 0 (S600: YES), S530
When M = 0 (S600: NO), S56
It is determined whether or not the difference between the individual R (M) obtained in the process of 0 or S580 is less than a predetermined value diff (S610).
If the difference of (M) is less than the predetermined value diff (S610: YE
S),), the average value of the individual R (M) is determined, and this is set as the radius of curvature (S620). Further, the difference of R (M) is a predetermined value.
If it is not less than diff (S610: NO), the radius of curvature estimated by the steering angle or the like obtained by the steering sensor 27 is used (S630).

On the other hand, if M ≦ N in the determination in S520 (S520: NO), the radius of curvature based on the target is not calculated, the process proceeds to S570, where M = 1, and then the process proceeds to S580. . In S580, the radius of curvature estimated from the steering angle or the like obtained by the steering sensor 27 is substituted for R (M). In this case, S580 to S59
When M is decremented after shifting to 0, M becomes 0, so a negative determination is made immediately in S600, and the flow shifts to S610.

As described above, when the vehicle is based on a moving object, if the moving object such as the own vehicle or the preceding vehicle changes lanes, a curvature different from that of an actual road will be estimated. , The moving objects larger than N are recognized (S520: YES), and the difference between the estimated values of the respective curvature radii is smaller than a predetermined threshold diff (S610: YES). Only, the radius of curvature based on the moving object is estimated (S620).

Returning to the flowchart of FIG. 4, in S90, a process of adding the estimation results is performed. This is achieved by performing, for example, a weighted average operation using the radius of curvature based on the steering angle estimated in S60 and the radius of curvature based on the recognized target estimated in S80, to obtain the final radius of curvature (R). This is the processing to be performed. Then, the preceding vehicle determination block 53 performs the lane determination of the recognized target based on the radius of curvature obtained by the addition processing as described above (S100).

Here, the lane determination processing in S100 will be described. In the present embodiment, the lane identity is determined based on the own lane probability of the recognition target. The own lane probability is a parameter indicating the likelihood that the target is a vehicle running on the same lane as the own vehicle. The instantaneous value of the own lane probability is a value calculated based on the detection data at that instant.

First, the positions of all targets obtained in the object recognition processing (S20) are converted into the positions when traveling on a straight road. Assuming that the original center position of the target is (Xo, Zo), a straight road conversion position (X, Z) is obtained by the following conversion formula (see FIG. 7A). X ← Xo−Zo ＾ 2 / 2R (Equation 8) Z ← Zo (Equation 9) R: Estimated R obtained in S90 of FIG. 4 Right curve: positive sign Left curve: negative sign “＾” means that the number before “＾” is raised to the power of the number after “＾”. The same applies to other parts of this specification. Here, the equation of the circle is | X
The approximation was performed under the assumption that | ≪ | R |, Z. Also,
When the radar device 5 is mounted at a position distant from the vehicle center, the X coordinate is corrected so that the vehicle center becomes the origin. That is, here, only the X coordinate is substantially converted.

The center position (X, Z) obtained as a result of the conversion to the straight road is placed on the own lane probability map shown in FIG. 8, and the instantaneous own lane probability of each object, that is, Calculates the probability of being in the own lane. The probability exists because the radius of curvature R obtained in S90 of FIG. 4 is a value estimated from the recognized target or the steering angle, and an error exists between the radius of curvature R and the actual radius of curvature of the curve. . In order to perform control in consideration of the error, an instantaneous own lane probability of each object is obtained here.

In FIG. 8, the horizontal axis is the X axis, that is, the horizontal direction of the own vehicle, and the vertical axis is the Z axis, that is, the front of the own vehicle. In the present embodiment, left and right 5 m, front 100 m
The area up to is shown. Here, the area is area a (own lane probability 80%), area b (own lane probability 60%), area c
(Own lane probability 30%), area d (own lane probability 100)
%), And other areas (own lane probability 0%). The setting of this area is determined by actual measurement.
In particular, the area d is an area set by taking into consideration the interruption immediately before the own vehicle.

Boundaries La, which delimit areas a, b, c, d,
Lb, Lc, and Ld are given by, for example, the following equations 10 to 13. Note that the boundary lines La ', Lb', L
c ′ and Ld ′ are boundary lines La, Lb, Lc and L, respectively.
d has a symmetrical relationship with the Y axis. La: X = 0.7 + (1.75-0.7) ・ (Z / 100) ＾ 2 [Equation 10] Lb: X = 0.7 + (3.5-0.7) ・ (Z / 100) ＾ 2 ... [Equation 11] Lc: X = 1.0 + (5.0-1.0) · (Z / 100) ＾ 2 [Equation 12] Ld: X = 1.5 · (1-Z / 60) [Equation 13] When this is expressed by a general equation, the following equation 14 is obtained. ~ 17. La: X = A1 + B1 · (Z / C1) ＾ 2 [Expression 14] Lb: X = A2 + B2 · (Z / C2) ＾ 2 [Expression 15] Lc: X = A3 + B3 · (Z / C3) ＾ 2 [Expression 16] Ld: X = A4 · (B4-Z / C4) [Expression 17] From Expressions 14 to 17, in general, the following Expressions 18 to 20 should be satisfied. Set the area. The actual numerical values are determined by experiments. A1 ≦ A2 ≦ A3 <A4 [Equation 18] B1 ≦ B2 ≦ B3 and B4 = 1 [Equation 19] C1 = C2 = C3 (C4 is not restricted) [Equation 20] The boundary line La in FIG. , Lb, Lc, La ', L
Although b 'and Lc' are parabolic from the viewpoint of calculation processing speed, if processing speed permits, it is better to represent them with arcs. The boundary lines Ld and Ld 'are preferably represented by parabolas or arcs bulging outward if the processing speed permits.

Next, the straight road conversion position of each target is compared with the own lane probability map of FIG. By collating with the map in the following manner, the own lane probability instantaneous value P0 can be obtained. An object having any area d → P0 = 100% An object having a center in area a → P0 = 80% An object having a center in area b → P0 = 60% An object having a center in area c → P0 = 30% An object that does not satisfy all of the above-> P0 = 0% Then, when the own lane probability instantaneous value is calculated for each target,
Next, filter processing is performed using the following equation. Where α
Is a parameter that depends on the distance Z, and is determined using the map of FIG. The initial value of the own lane probability is set to 0%. Own lane probability ← previous own lane probability × α + instantaneous lane probability instantaneous value × (1-α) Then, the preceding vehicle is determined using the own lane probability. Specifically, among the targets having the own lane probability of 50% or more, the target having the minimum distance Z is determined to be the preceding vehicle. The result of this determination, that is, the recognized target determined to be the preceding vehicle and the own lane probability thereof are output to the inter-vehicle control unit and the warning determination unit block 55 (S110).

In this embodiment, the radar device 5 corresponds to radar means, and the object recognition block 43 and the road curvature estimation block 45 correspond to curvature estimation means. Of these, the object recognition block 43 corresponds to the object recognition means,
The road curvature estimation block 45 corresponds to a change rate calculation unit, a curvature calculation unit, and an average value calculation unit. At least one set of the steering sensor 27, the steering angle calculation block 49, the yaw rate sensor 28, and the yaw rate calculation block 51 corresponds to turning detection means, and the curvature radius calculation block 6
3 corresponds to a second curvature estimating means. Further, the preceding vehicle determination block 53 corresponds to own lane probability calculating means and preceding vehicle selecting means.

As described above, in the present embodiment, based on the past history of the horizontal position of the recognized target, the change with time of the horizontal position of the object is approximated by a straight line, and the change rate is calculated. The curvature of the road ahead of the vehicle is estimated based on the calculated change rate. Therefore, even if the vehicle is traveling on a straight line but curves ahead,
It can properly estimate the curvature of the road ahead of it,
The preceding vehicle can be appropriately selected based on the estimated road curvature.

The radius of curvature based on the recognition target is estimated only when the predetermined condition in S70 of FIG. 4 is satisfied (at least when the first and second data are continuously detected). Therefore, more appropriate curvature estimation can be performed. In the present embodiment, when a stationary object is recognized (YES in S320 of FIG. 5), the curvature radius is preferentially estimated based on the stationary object. Although the curvature of the road can be estimated for either a stationary object or a moving object, a stationary object is preferable. Because, in the case of a moving object, its speed may change or the lane may change, and furthermore, the traveling position may move left and right even in the same lane, so if it is a stationary object, such This is because it is not necessary to consider an estimation error due to a significant change.

In the above-described embodiment, in the lane determination process in S100 of FIG. 4, the identity of the lane is determined based on the own lane probability of the recognition target. On the other hand, it is also possible to determine lane identity based on the radius of curvature obtained based on the recognized target. This will be described as a second embodiment.

The configuration of the second embodiment is the same as that of FIG. 1, except that the processing executed in the road curvature estimation block 45 and the preceding vehicle determination block 53 is different. FIG.
Is a flowchart showing processing in the case of the second embodiment. In this processing, S1010 to S1040 are the same processing as S10 to S40 in FIG.
The description starts from 1050.

In step S1050, the road curvature is estimated based on the recognized target. This process will be described with reference to the flowchart of FIG. In FIG. 11, the relative speed V, the lateral position X, and the distance D are extracted as data of the recognized target (S1310), and the number of recognized targets is set to M (S132).
0), R for storing the radius of curvature determined in a later process
[M] is initialized (S1330). And S1340
To S1380 are repeated M times. More specifically, the number of consecutive recognitions exceeds n (S1340: Y
ES), and if the recognition start distance is longer than d (S1350: YES), the flow shifts to S1360; otherwise, the flow shifts to S1380. In S1360,
Based on the history of the lateral position X _M (t) data, the linear equation X _M (t) = a
Linear approximation is performed using t + b to determine the slope a, and the following S
At 560, the distance D and the relative speed V of the target history data
And the slope a of the straight line obtained in S550, the curvature radius R [M] of the road is obtained by R [M] = DV / a. When the processing of S1340 to S1380 has been repeated M times (S1390: NO), this processing routine ends, and the flow shifts to S1060 in FIG.

In S1060, the lane of the recognized target is determined. This process will be described with reference to the flowchart of FIG. First, the curvature radius R [M] obtained from the stationary object is extracted (S1510). And the extracted R
The number of [M] is set to L (S1520), and if L <2 (S1530: NO), the process proceeds to S1700. S17
In the case of shifting to 00, since L is 1 or 0, that is, there are no more than two stationary objects, it is not possible to determine the same lane by comparing the curvature radii of the recognized targets. Therefore, in S1700, a determination using the own lane probability described in S100 of FIG. 4 in the first embodiment is performed using the average value of the individual radii of curvature R [M] extracted in S1510, and this processing is performed. To end.

On the other hand, if L ≧ 2 (S1530: YE
S), the extracted radius of curvature R [M] is substituted into the radius of curvature Rs [L] based on the stationary object (S1540). The one having the largest absolute value among the curvature radii Rs [L] based on the stationary object is set to Rsmax (S1550), and the one having the smallest absolute value among Rs [L] is set to Rsmin (S1560).

At S1570, it is determined whether or not Rsmax-Rsmin is larger than a predetermined threshold value diffR.
If ax-Rsmin> diffR (S1570: YES),
The flow shifts to S1580, where the process of lane identity determination using the radius of curvature of the recognized target is continued. On the other hand, Rsmax−Rsm
If in ≦ diffR (S1570: NO), S1700
Then, the determination using the own lane probability is performed. S157
When such a determination is made at 0, the individual radii of curvature of the recognized stationary object are compared, and if the difference between them is greater than a certain value, it can be determined that the stationary object exists at both ends of the road. Only in this case, the same lane determination using the radius of curvature of the moving object is performed.

In step S1580, the object for which Rs [L], which is Rsmax, has been obtained, that is, Rs obtained from the recognized stationary object.
The latest data of the lateral position X of the stationary object such that [M] becomes Rsmax is defined as RXsmax. Similarly, in S1590, Rsmin
Let RXsmin be the target for which Rs [L] is obtained, that is, the latest data of the horizontal position X of the stationary object such that R [M] obtained from the recognized stationary object is Rsmin. Then, the curvature radius R [M] obtained from the moving object is extracted (S160).
0), and the number of the extracted R [M] is set to K (S16).
10). Further, the retrieved radius of curvature R [M] is substituted into the radius of curvature Rm [K] based on the stationary object (S1620).

In the following S1630, the R obtained in S1580
It is determined whether the absolute value of Xsmax is greater than the absolute value of RXsmin obtained in S1590. The meaning of this determination will be described. The judgment formula itself compares the magnitude of the lateral position of the recognized stationary object with reference to the traveling direction of the own vehicle. for that reason,
When | RXsmax |> | RXsmin |, it is considered that the own vehicle is traveling on the lane having a smaller radius of curvature (for example, the left lane in the case of a left curve). In the opposite case, your vehicle
It is considered that the vehicle is traveling on the lane with the larger radius of curvature (for example, the lane on the right side in the case of a left curve). Therefore,
One of the stationary objects at both ends of the road is used as a criterion for determining the same lane. Specifically, | RXsmax |> | RXsmin, based on a stationary object at the road end closer to the own vehicle.
| When | RXsmin |, | RXsmax | ≦ | RXsmin |
In this case, | RXsmax | is used as a reference.

Therefore, when | RXsmax |> | RXsmin | (S1630: YES), Rm obtained in S1620
The value obtained by subtracting the absolute value of RXsmin, which is the reference in this case, from the absolute value of [K] is less than a predetermined threshold value diffX (| Rm
It is determined whether [K] | − | RXsmin | <diffX) (S
1640). If this condition is satisfied (S16
40: YES), it is determined that the recognized object used for obtaining Rm [K] is present on the own lane (S1660), and if the condition is not satisfied (S1640: NO), it is determined that Rm [K] is obtained. It is determined that the used recognition object does not exist on the own lane (S1670). This is because if the difference between the radius of curvature Rs obtained from the reference stationary object and the radius of curvature Rm obtained from the moving object to be determined is within a certain range, the moving object is considered to be on the own lane. Because it can be done.

On the other hand, when | RXsmax | ≦ | RXsmin | (S1630: NO), the reference RXsm
A value obtained by subtracting the absolute value of Rm [K] from the absolute value of ax is smaller than a predetermined threshold value diffX (| RXsmax |-| Rm [K] | <
diffX) (S1650). If this condition is satisfied (S1650: YES), Rm
It is determined that the recognized object used for obtaining [K] exists on the own lane (S1660), and if the condition is not satisfied (S1660).
1650: NO), it is determined that the recognized object used for obtaining Rm [K] does not exist on the own lane (S1670).

After the processing of S1660 or S1670, K
Is decremented (S1680). Such S16
The processing from 30 to S1680 is repeated until K becomes 0, and when K becomes 0 (S1690: YES),
This processing ends. In the second embodiment,
The preceding vehicle determination block 53 corresponds to a lane identicalness determination unit.

As described above, in the second embodiment, it is possible to perform the same lane determination based on the road curvature obtained from the recognition object instead of the own lane probability as in the first embodiment. Yes, you can select the preceding car.
It should be noted that the present invention is not limited to such embodiments at all, and can be implemented in various forms without departing from the gist of the present invention.

For example, in the above embodiment, the radar device 5 using a millimeter wave or the like is employed as the "radar means", but a laser radar device using a laser beam may be employed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a vehicle control device to which the present invention has been applied.

FIG. 2 is an explanatory diagram of a lateral position change based on a locus of a recognition target on a curved road.

FIG. 3 is an explanatory diagram showing that a curvature radius of a road can be appropriately obtained based on a trajectory of a recognition target.

FIG. 4 is a flowchart illustrating an outline of a lane identification process.

FIG. 5 is a flowchart showing a road curvature estimation process executed in FIG. 4;

FIG. 6 is a flowchart showing a process for a moving object executed in FIG. 5;

FIG. 7A is an explanatory diagram when converting each target position into a position on a straight road, and FIG. 7B is an explanatory diagram of a map of a parameter α for obtaining the own lane probability.

FIG. 8 is an explanatory diagram of an own lane probability map.

FIG. 9 is an explanatory diagram showing that lane identity of a moving object is appropriately determined based on a radius of curvature of a road obtained from a recognition target.

FIG. 10 is a flowchart illustrating an outline of a lane identity determination process according to the second embodiment.

11 is a flowchart illustrating a road curvature estimation process executed in FIG. 10;

12 is a flowchart illustrating a lane determination process for a recognized target executed in FIG. 10;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Vehicle control apparatus 3 ... Computer 5 ... Radar apparatus 5a ... Transmission / reception part 5b ... Signal processing part 7 ... Vehicle speed sensor 9 ... Brake switch 11 ... Throttle opening degree sensor 13 ... Alarm sound generator 15 ... Distance display 17 ... Sensor abnormality Indicator 19 ... Brake driver 21 ... Throttle driver 23 ... Automatic transmission controller 24 ... Alarm volume setting device 25 ... Alarm sensitivity setting device 26 ... Cruise control switch 27 ... Steering sensor 28 ... Yaw rate sensor 29 ... Power switch 41 ... Polar / Cartesian coordinate conversion block 43 ... Object recognition block 45 ... Road curvature estimation block 47 ... Vehicle speed calculation block 49 ... Steering angle calculation block 51 ... Yaw rate calculation block 53 ... Preceding vehicle judgment block 55 ... Vehicle control unit and alarm judgment unit block 63 … Curvature radius calculation block

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) B60R 21/00 624 B60R 21/00 624D 626 626B 626E F02D 29/02 301 F02D 29/02 301D F term (reference) 3D044 AA21 AA25 AB01 AC03 AC24 AC31 AC56 AC59 3G093 AA01 BA14 BA23 BA24 BA27 CB09 CB10 DA06 DB05 DB15 DB16 DB18 EA09 EB03 EB04 EC01 FA02 FA04 FA11 5H180 AA01 CC14 LL01 LL06 LL07 LL09 LL15

Claims (11)

[Claims] 1. A radar means for irradiating a transmission wave over a predetermined angle range in the vehicle width direction and detecting an object based on the reflected wave, and a road curvature ahead of the vehicle based on a detection result by the radar means. A curvature estimating device for a vehicle, comprising: a curvature estimating means for estimating, wherein the curvature estimating means is based on a detection result by the radar means, at least a distance to an object and an extension of a longitudinal axis of the own vehicle. Object recognition means for calculating an object lateral position, which is a physical quantity indicating how far the object is in the vehicle width direction when the object is referred to, and an object lateral position calculated by the object recognition means Change rate calculating means for calculating a change rate of the object lateral position from a history with the passage of time, based on the change rate calculated by the change rate calculating means,
A curvature estimating device for a vehicle, comprising: a curvature calculating means for calculating a curvature of a road ahead of the vehicle. 2. The vehicle road curvature estimating apparatus according to claim 1, wherein the change rate calculating means calculates the change rate by approximating a history of the lateral position of the object over time with a straight line. Vehicle road curvature estimating device. 3. The road curvature estimating device for a vehicle according to claim 1, wherein the change rate calculating means detects at least first and second data successively at an initial stage of detecting the object. A road curvature estimating device for a vehicle, which calculates a rate of change only when there is a vehicle. 4. The road curvature estimating device for a vehicle according to claim 1, wherein the object recognizing means also determines a recognition type as a moving object or a stationary object based on the relative speed of the object and the own vehicle speed. It is possible to determine, the change rate calculating means, when the stop object is recognized, calculates a change rate using the object lateral position of the stop object, the moving object is not recognized the stop object A road curvature estimating device for a vehicle, which calculates a rate of change using a lateral position of the moving object when the moving object is recognized. 5. The vehicle road curvature estimating device according to claim 1, further comprising an average value calculating means for calculating an average value of a predetermined number of curvatures calculated by said curvature calculating means. A road curvature estimating device for a vehicle. 6. The vehicle road curvature estimating device according to claim 5, wherein the turning detecting means for detecting a turning state of the own vehicle, and the turning state of the own vehicle and the own vehicle speed detected by the turning detecting means. Second curvature estimating means for estimating the curvature of the road based on the second curvature when the condition that the change rate calculating means can calculate the change rate is not satisfied. An apparatus for estimating a road curvature for a vehicle, comprising: adopting a curvature estimated by an estimating means and calculating an average value of the predetermined number of times including the curvature. 7. A vehicle road curvature estimating apparatus according to claim 1, wherein said vehicle detecting means detects a turning state of said own vehicle, and said vehicle turning state detected by said turning detecting means. A second curvature estimating means for estimating the curvature of the road based on the vehicle speed and the vehicle speed based on the curvature estimated by the second curvature estimating means and the curvature estimated by the curvature estimating means. A vehicle road curvature estimating device for estimating a final road curvature. 8. The vehicle road curvature estimating device according to claim 1, wherein the object is determined based on a road curvature obtained by the vehicle road curvature estimating device and a relative position of the object. A preceding lane comprising: an own lane probability calculating means for calculating a probability of being on the same lane as the own vehicle; and a preceding vehicle selecting means for selecting a preceding vehicle based on the probability obtained by the own lane probability calculating means. Car selection device. 9. A road curvature estimating device for a vehicle according to any one of claims 1 to 7, and the road curvature estimating device for a vehicle based on stationary objects assumed to be present at both ends of the road. Is obtained, based on the magnitude relationship between the road curvature and the road curvature obtained based on the moving object, it is determined that there is a possibility that the moving object is on the same lane as the own vehicle. A preceding vehicle selecting device, comprising: determining means; and preceding vehicle selecting means for selecting a preceding vehicle based on a result of the determination by the lane identity determining means. 10. The preceding vehicle selecting device according to claim 9, wherein the difference between the radius of curvature of the road is a predetermined value, based on whether the road curvature is based on a stationary object assumed to exist at both ends of the road. A preceding vehicle selecting device, wherein the determination is made based on whether or not the above is true. 11. The preceding vehicle selecting device according to claim 9, wherein the same lane judging means determines whether the own vehicle is left or right on the basis of an object lateral position of a stationary object assumed to be present at both ends of the road. It is determined which of the road ends is closer to the road end, the result of the determination, the road curvature based on the moving object to be determined whether or not the lanes are the same, and the stop assumed to be present at both ends of the road A preceding vehicle selecting apparatus for determining the lane identity of a moving object based on a magnitude relationship with a road curvature based on the object.