JP5904226B2 - Vehicle behavior prediction apparatus and program - Google Patents

Description

  The present invention relates to a vehicle behavior prediction apparatus and program, and more particularly, to a vehicle behavior prediction apparatus and program that predicts position information and a motion state of a vehicle.

  Conventionally, on the basis of the relationship between the road condition and the acceleration of the vehicle obtained in advance for each of a plurality of types of operation conditions, the road condition at each point, and the time series data of the predicted operation condition, There is known a speed prediction device that predicts time-series data of acceleration and predicts time-series data of the speed of the vehicle on the traveling road based on the time-series data of the acceleration of the vehicle on the traveling road (Patent Document 1).

JP 2012-168037 A

  However, since the technique described in Patent Document 1 uses information that cannot be observed from outside the vehicle, such as vehicle speed, acceleration, and operation amount, in addition to the vehicle position information, the apparatus is mounted. It cannot be used to predict the vehicle that is being used.

  In addition, since a plurality of driving scenes are not assumed, the prediction accuracy of general road driving is low.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a vehicle behavior prediction apparatus and program capable of accurately predicting position information and a motion state of a vehicle.

In order to achieve the above object, a vehicle behavior prediction apparatus according to the present invention includes an acquisition unit that acquires position information of a vehicle and external environment information of the vehicle at each time t, and a time t− acquired by the acquisition unit. A driving operation prediction unit that predicts a driving operation amount of the driver of the vehicle at time t based on the external environment information of the vehicle at 1 and the motion state of the vehicle at time t−1, and the driving operation prediction unit. Vehicle behavior prediction for predicting the position information of the vehicle and the motion state of the vehicle at the time t based on the driving operation amount of the driver at the time t predicted by the time t and the motion state of the vehicle at the time t−1. And the vehicle position information at time t acquired by the acquisition means and the vehicle position information at time t predicted by the vehicle behavior prediction means. Driving operation vehicle state estimation means for estimating the movement state of the vehicle, wherein the driving operation prediction means includes the external environment information of the vehicle at time t acquired by the acquisition means, and the driving operation vehicle state estimation means Based on the estimated movement state of the vehicle at the time t, the driving operation amount of the driver of the vehicle at the time t + 1 is predicted, and the vehicle behavior prediction unit is configured to predict the driving operation prediction unit at the time t + 1 predicted by the driving operation prediction unit. Based on the driving operation amount of the driver and the movement state of the vehicle at time t estimated by the driving operation vehicle state estimation means, the position information of the vehicle and the movement state of the vehicle at time t + 1 are predicted.

  The program according to the present invention includes a computer for acquiring vehicle position information and vehicle external environment information for each time t, and vehicle external environment information at time t−1 acquired by the acquisition device. And the driving operation predicting means for predicting the driving operation amount of the driver of the vehicle at the time t based on the motion state of the vehicle at the time t−1, the driver at the time t predicted by the driving operation predicting means. Obtained by the vehicle behavior prediction means for predicting the position information of the vehicle and the motion state of the vehicle at the time t, and the acquisition means based on the driving operation amount of the vehicle and the motion state of the vehicle at the time t-1. Based on the position information of the vehicle at time t and the position information of the vehicle at time t predicted by the vehicle behavior prediction means, the motion state of the vehicle at time t is estimated. The driving operation predicting means is estimated by the driving operation vehicle state estimating means and the external environment information of the vehicle at time t acquired by the acquiring means. Based on the movement state of the vehicle at the time t, the driving operation amount of the driver of the vehicle at the time t + 1 is predicted, and the vehicle behavior predicting unit predicts the driver's driving operation at the time t + 1 predicted by the driving operation prediction unit. The program predicts the position information of the vehicle and the movement state of the vehicle at time t + 1 based on the driving operation amount and the movement state of the vehicle at time t estimated by the driving operation vehicle state estimation means.

  According to the present invention, the acquisition unit acquires the position information of the vehicle and the external environment information of the vehicle for each time t.

  Then, the driving operation amount of the driver of the vehicle at time t based on the external environment information of the vehicle at time t-1 acquired by the acquisition unit and the motion state of the vehicle at time t-1 by the driving operation prediction unit. Predict.

  Then, based on the driving operation amount of the driver at the time t predicted by the driving operation prediction unit and the motion state of the vehicle at the time t−1, the vehicle behavior prediction unit and the vehicle position information and the vehicle Predict movement status.

  Then, based on the position information of the vehicle at time t acquired by the acquisition means and the position information of the vehicle at time t predicted by the vehicle behavior prediction means by the driving operation vehicle state estimation means, the vehicle at time t Estimate the state of movement.

  Further, the driving operation predicting means is based on the external environment information of the vehicle at the time t acquired by the acquiring means and the motion state of the vehicle at the time t estimated by the driving operation vehicle state estimating means. Predict the amount of driver operation.

  Further, the vehicle behavior predicting means is based on the driving operation amount of the driver at time t + 1 predicted by the driving operation predicting means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimating means. Vehicle position information and vehicle motion state are predicted.

  Thus, based on the external environment information of the vehicle and the position information of the vehicle, the position information and the motion state of the vehicle can be accurately predicted.

  According to the present invention, the driving operation prediction means includes the external environment information of the vehicle at time t acquired by the acquisition means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. Based on the probability distribution, the probability distribution of the driving operation amount of the driver of the vehicle at the time t + 1 is predicted, and the vehicle behavior prediction unit is configured to determine the driving operation amount of the driver at the time t + 1 predicted by the driving operation prediction unit. Based on the probability distribution and the probability distribution of the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means, the probability distribution of the position information of the vehicle at time t + 1 and the probability of motion state of the vehicle The driving operation vehicle state estimating means predicts the distribution, and the vehicle position information obtained at the time t + 1 acquired by the acquiring means and the vehicle behavior predicting means. Based on the probability distribution of the location information of the vehicle of the expected time t + 1, the probability distribution of the motion state of the vehicle at time t + 1 can be made to estimate.

  In the present invention, an upper limit speed that satisfies each of a plurality of predetermined speed constraint conditions is calculated based on the external environment information, and any one of the calculated upper limit speeds is selected as a target speed. The driving operation prediction unit further includes a target speed selection unit, and the driving operation prediction unit includes the target speed selected by the target speed selection unit, the external environment information of the vehicle at time t acquired by the acquisition unit, and the driving operation. Based on the motion state of the vehicle at time t estimated by the vehicle state estimation means, the driving operation amount of the driver of the vehicle at time t + 1 can be predicted.

  In addition, the present invention further includes target position determination means for determining a target position on the travel path of the vehicle for each time t, and the acquisition means is a travel path on which the vehicle travels as external environment information of the vehicle. A steady traveling speed, a curvature of a traveling path on which the vehicle travels, and forward vehicle information indicating a distance from a vehicle traveling in front of the vehicle are acquired at each time t, and the target speed selecting means A traveling speed is calculated as an upper limit speed that satisfies a speed constraint condition regarding steady traveling, an upper limit speed that satisfies a speed constraint condition regarding curve traveling is calculated based on the curvature of the traveling road, and a forward vehicle is calculated based on the preceding vehicle information. An upper limit speed that satisfies a speed constraint condition for tracking is calculated, and a speed for front-rear position control is calculated based on the target position on the track determined by the target position determining means. Calculated approximately satisfy the upper limit speed, any of the calculated maximum speed may be adapted to select as the target speed.

  The acquisition means according to the present invention further acquires signal information on the road of the vehicle and a destination of the vehicle for each time t as the external environment information of the vehicle, and the target position determination means. Calculates a vehicle position that satisfies a predetermined position constraint condition regarding signal stop based on the signal information, and sets a position constraint condition regarding the route based on the route information obtained based on the destination of the vehicle. It is possible to calculate a vehicle position to be satisfied, select one of the calculated vehicle positions, and determine the selected position as a target position.

  Further, the driving operation prediction means in the present invention includes the external environment information of the vehicle at time t acquired by the acquisition means, the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means, and Based on the driving operation amount of the driver of the vehicle, the driving operation amount of the driver of the vehicle at time t + 1 is predicted, and the driving operation vehicle state estimation means obtains the position information of the vehicle at time t + 1 acquired by the acquisition means. And the position of the vehicle at time t + 1 predicted by the vehicle behavior predicting means, the motion state of the vehicle at time t + 1 and the driving operation amount of the driver of the vehicle can be estimated. .

  The program of the present invention can also be provided by being stored in a storage medium.

  As described above, according to the speed prediction apparatus and program of the present invention, based on the acquired vehicle position information at time t and the vehicle position information predicted at time t, The movement state is estimated, and the driving operation amount of the driver of the vehicle at time t + 1 is predicted based on the external environment information of the vehicle at time t and the estimated movement state of the vehicle at time t. Based on the driving amount of the driver and the estimated motion state of the vehicle at time t, by predicting the vehicle position information and motion state at time t + 1, the external environment information of the vehicle, the vehicle position information, Based on the above, it is possible to accurately predict the position information and the motion state of the vehicle.

1 is a block diagram showing a vehicle behavior prediction apparatus according to a first embodiment of the present invention. It is a figure which shows the state space model which concerns on the 1st Embodiment of this invention. It is a conceptual diagram which shows the concept of the prediction process in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a conceptual diagram which shows the concept of the filtering process in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the detail of the driving operation prediction part 26 of the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the detail of the vehicle behavior prediction part 28 of the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the vehicle behavior prediction process routine in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the prediction process routine in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the driving operation prediction process routine in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the vehicle prediction process routine in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the driving operation vehicle state estimation process routine in the vehicle behavior prediction apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the vehicle behavior prediction apparatus which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the detail of the driving operation prediction part 226 of the vehicle behavior prediction apparatus which concerns on the 2nd Embodiment of this invention.

  Embodiments of the present invention will be described below with reference to the drawings. An example will be described in which the present invention is applied to a vehicle behavior prediction apparatus that estimates a driving operation amount and a motion state based on vehicle position information and vehicle external environment information and predicts the behavior of the vehicle. In the present embodiment, an example will be described in which an extended Kalman filter is used to estimate the driving operation amount and the motion state and predict the behavior of the vehicle, assuming that the prediction target vehicle is the own vehicle.

  As shown in FIG. 1, the vehicle behavior prediction apparatus 10 according to the first embodiment stores an input operation unit 12 for a driver to input a destination, map information (road network data), and road information. The road network database 14, the forward vehicle information acquisition unit 16 that detects a forward vehicle traveling ahead of the prediction target vehicle, the position measurement unit 18 that measures the position of the prediction target vehicle, and the computer 20 that predicts the behavior of the vehicle. And an output device 40 that outputs the prediction result.

  The road network database 14 stores map information and road information. The road information includes road speed information and road curvature information. The travel speed information includes information on the steady travel speed on the travel path. The steady travel speed is calculated from past travel data and a speed limit on the travel path. The track curvature information includes information related to the curvature of the track.

  The forward vehicle information acquisition unit 16 corresponds to, for example, a laser radar device, measures the relative distance in the traveling direction to the forward vehicle with respect to the prediction target vehicle, and uses the relative distance and the relative speed of the forward vehicle as the forward vehicle information at each time. Get for t. The relative speed is obtained by differentiating the relative distance with respect to time.

  The position measurement unit 18 is configured using, for example, a GPS sensor, and measures the position of the prediction target vehicle at each time t.

  The computer 20 stores a CPU, a RAM, and a program for executing a vehicle behavior prediction processing routine, a prediction processing routine, a driving operation prediction processing routine, a vehicle prediction processing routine, and a driving operation vehicle state estimation processing routine, which will be described later. A ROM is provided and is functionally configured as follows. The computer 20 acquires the destination information input by the input operation unit 12, maps the map information and road information obtained from the road network database 14, and forward vehicle information detected by the forward vehicle information acquisition unit 16, Measured by the vehicle information acquisition unit 22 acquired as the external environment information of the vehicle, the prediction unit 24 that predicts the driving operation amount of the driver, the vehicle position information, and the motion state based on the acquired external environment information, and the position measurement unit 18 Based on the positional information acquisition unit 30 that acquires the positional information of the vehicle that has been performed, the positional information acquired by the positional information acquisition unit 30, and the prediction result predicted by the prediction unit 24, and And a driving operation vehicle state estimating unit 32 for estimating the motion state of the vehicle. The process performed by the prediction unit 24 corresponds to the prediction step in the extended Kalman filter, and the process performed by the driving operation vehicle state estimation unit 32 corresponds to the filtering step in the extended Kalman filter. The vehicle information acquisition unit 22 and the position information acquisition unit 30 are examples of acquisition means.

The vehicle information acquisition unit 22 acquires the external environment information of the vehicle for each time t.
Specifically, when the vehicle information acquisition unit 22 predicts the behavior of the vehicle, the vehicle information acquired by the vehicle information acquisition unit 16 ahead, the map information and the road at the position obtained from the road network database 14 Information is sequentially acquired and accumulated as external environment information of the vehicle. Further, the vehicle information acquisition unit 22 acquires the destination information received by the input operation unit 12 for each time t.

  In the present embodiment, the driver's driving operation amount, the vehicle position information and the motion state are predicted using the extended Kalman filter, and the driver's driving operation amount and the motion state are estimated. Hereinafter, the principle of the extended Kalman filter will be described.

In the present embodiment, the vehicle position [x _t , y _t ], the vehicle orientation θ _t , the vehicle speed v _t , the vehicle acceleration α _t, and the vehicle curvature δ _t are _expressed by the following equations: As shown in (1), a state vector X _t = [x _t y _t θ _t v _t α _t δ _t ] is constructed. Further, as shown in the following formula (2), the position information of the vehicle measured by the position measuring unit 18 is configured as an observation vector Y _t = [x _t ^* y _t ^* ].

(1) Prediction Step First, an outline of processing in the prediction step will be described. In the prediction step, the state vector X _{t at} time _t is calculated based on the state vector X _{t−1 at} time _t−1 and the system noise u _t (error term) as shown in the following equation (3). To do. In the following equation (3), _ft is a nonlinear function, and in the present embodiment, it is expressed by a driver model and a vehicle physical model as shown in FIG. The driver model in FIG. 2 corresponds to the driving operation prediction unit 26, and the vehicle physical model corresponds to the vehicle behavior prediction unit 28.

The observation vector Y _t is calculated as shown in the following equation (4) based on the observation matrix H _t and the observation noise w _t (error term). Note that the observation matrix H _t is obtained in advance.

FIG. 3 shows a conceptual diagram of the prediction step. As shown in FIG. 3, in the prediction step, the state vector X _t-1 at time t-1 which is estimated by the filtering step _| based on _t-1, the state vector X _t at time t _| predicting _t-1 (Predict the center value). X _{t | t−1} represents a state vector at time t predicted based on data available up to time t−1. X _{t−1 | t−1} represents a state vector at time t−1 estimated based on data available up to time t−1.

  In the following description, the value to which the subscript t | t−1 is given represents the prior estimated value of the time t predicted based on the data available up to the time t−1. Further, the value to which the subscript t | t (or t-1 | t-1) is given represents the post-estimation value at time t estimated based on the data available up to time t.

In the prediction step, a matrix ^ F _t and a matrix ^ G _t obtained by partial differentiation of the function f _t are calculated according to the following equations (5) to (6).

  Here, “′” represents transposition of the vector.

Then, the equation (5) to on the basis of the calculated matrix ^ _{F t} and the matrix ^ _{G t} according (6), according to the following equation (7), pre-error covariance matrix _{P t |} calculating the _t-1 (Distribution prediction).

In addition, Q _t in the above equation (7) represents the variance of the system noise u _t (error term).

That is, in the prediction step, a probability distribution having the state vector X _{t | t−1} at time t as the center value and the prior error covariance matrix P _{t | t−1} as the variance is predicted.

(2) Filtering Step Next, an outline of the filtering step will be described. FIG. 4 shows a conceptual diagram of the filtering step. As shown in FIG. 4, in the filtering step, the vehicle position [x _t , y _t ], which is an element of the state vector X _{t at} the time t shown in the equation (1), and the equation (2) Filtering is performed based on the position information [x ^* _t , y ^* _t ] (observation center) of the vehicle at time t, and the state vector _{Xt | t-1} output at the prediction step is corrected. X _{t | t} is calculated, and the prior error covariance matrix P _{t | t−1} is modified to calculate the posterior error covariance matrix P _{t | t} .

Specifically, first, in the filtering step, the prior error covariance matrix P _{t | t−1} calculated in the prediction step, the observation matrix H _t, and the variance R _{t of} the observation noise w _t (error term). Based on the following equation (8), the Kalman gain K _t is calculated.

Next, the calculated Kalman gain K _t , the state vector X _{t | t−1} calculated at the time t calculated in the prediction step, and Y _t = [x ^* _t , y ^* _t shown in the above equation (2). ] And the observation matrix H _t , the state vector X _{t | t} at time _t is estimated according to the following equation (9).

Then, based on the prior error covariance matrix P _{t | t−1} calculated in the prediction step, the calculated Kalman gain K _t, and the observation matrix H _t , the posterior error covariance is calculated according to the following equation (10). A variance matrix P _{t | t} is calculated.

Each value (Kalman gain K _t , state vector X _{t | t} , posterior error covariance matrix P _{t | t} ) calculated in the filtering step is used in processing in the next prediction step.

The prediction unit 24 includes the external environment information of the vehicle at time t-1 acquired by the vehicle information acquisition unit 22, the probability distribution of the motion state of the vehicle at time t-1 estimated by the driving operation vehicle state estimation unit 32, and Based on the probability distribution of the driving operation amount of the driver, the probability distribution of the driving operation amount of the driver at time t, the probability distribution of the vehicle position information, and the probability distribution of the vehicle motion state are predicted. In the present embodiment, as the driver's driving operation amount, the vehicle curvature δ _t based on the steering operation amount and the vehicle acceleration α _t based on the accelerator pedal operation amount are predicted. Further, as the vehicle position information and the motion state, the vehicle position [x _t , y _t ], the vehicle orientation θ _t, and the vehicle speed v _t are predicted.

  In the present embodiment, vehicle position information is acquired every 0.1 second. In addition, the prediction unit 24 includes the external environment information of the vehicle 0.1 seconds before (time t−1) acquired by the vehicle information acquisition unit 22 and 0.1 seconds estimated by the driving operation vehicle state estimation unit 32. Based on the previous state of motion (time t-1) of the vehicle and the amount of driving operation of the driver, the amount of driving operation of the driver at the current time (time t), vehicle position information, and the state of motion of the vehicle are predicted. The prediction unit 24 repeats the driving operation amount prediction process of the driver in the driving operation prediction unit 26 and the vehicle position information and the motion state prediction process of the vehicle behavior prediction unit 28 30 times for 3 seconds. It is assumed that the position information and the motion state of the vehicle until later are predicted.

Specifically, the prediction unit 24 includes a driving operation prediction unit 26 and a vehicle behavior prediction unit 28. Driving operation prediction unit 26, the vehicle information acquisition unit and the external environment information of the time t-1 of the vehicle obtained by 22, the driving operation vehicle state estimating unit time t-1 of the estimated by 32 state vector X _t-1 Based on (the vehicle position information, the motion state, and the driving operation amount of the driver), the driving operation amount of the driver at time t is predicted. As shown in FIG. 5, the driving operation prediction unit 26 includes a driving operation reception unit 100, a route generation unit 102, a target position selection unit 104, a lateral position control unit 106, a target speed restriction unit 108, and a target speed. A selection unit 118, a speed control unit 120, and a driving operation output unit 122 are provided.

The driving operation reception unit 100 receives the state vector X _t-1 at time t−1 estimated by the driving operation vehicle state estimation unit 32.

The route generation unit 102 is an element of the destination vector and map information acquired by the vehicle information acquisition unit 22 and the state vector X _t−1 of the time t−1 received by the driving operation reception unit 100 [x _{t -1} , y _t-1 ] based on the shortest route, the shortest route from the starting point to the destination using an algorithm for searching for the shortest route (for example, A * algorithm) Is generated.

The target position selection unit 104 determines a target position on the traveling path of the vehicle for each time t based on the shortest route generated by the route generation unit 102. For example, the target position selection unit 104 includes the route information obtained from the destination of the vehicle generated by the route generation unit 102 and the element of the state vector X _{t-1 at} the time t−1 received by the driving operation reception unit 100. Based on [x _t−1 , y _t−1 ] that is, a point 100 m ahead from the position [x _t−1 , y _t−1 ] on the route is selected as the target position.

The lateral position control unit 106 is an element of the target position [x _r , y _r ] selected by the target position selection unit 104 and the state vector X _t−1 at time t−1 received by the driving operation reception unit 100. Based on [x _t−1 , y _t−1 , v _t−1 , θ _t−1 ], the steering operation amount of the driver is predicted. Then, the lateral position control unit 106 calculates the curvature δ of the vehicle based on the predicted steering operation amount. The relationship between the steering operation amount and the curvature δ is obtained in advance. For example, the lateral position control unit 106 predicts the steering operation amount using the forward gaze model, and calculates the curvature δ based on the predicted steering operation amount.

The target speed restriction unit 108 includes the external environment information of the vehicle at time t−1 acquired by the vehicle information acquisition unit 22, the target position [x _r , y _r ] selected by the target position selection unit 104, and the driving operation. based on the state vector X _t-1 at time t-1 which received by the receiving unit 100, calculates the upper limit speed which satisfies each of the plurality of speed constraints. Each of the plurality of speed constraint conditions corresponds to various scenes in the driving operation. The target speed constraint unit 108 includes a front / rear position control constraint unit 110, a steady travel constraint unit 112, a follow-up constraint unit 114, and a curve constraint unit 116.

The front / rear position control restriction unit 110 uses [x _t−1 , y _t−1 ], which are elements of the state vector X _t−1 at time t−1 received by the driving operation reception unit 100, and the target position selection unit 104. Based on the selected target position [x _r , y _r ] on the running road, an upper limit speed that satisfies a speed constraint condition regarding front-rear position control is calculated. For example, the front / rear position control restriction unit 110 uses a proportional control model to calculate the difference between the vehicle position [x _t−1 , y _t−1 ] at time _t−1 and the target position [x _r , y _r ]. Accordingly, an upper limit speed is calculated.

The steady travel restriction unit 112 is acquired by [x _t−1 , y _t−1 ], which is an element of the state vector X _{t−1 at} the time t−1 received by the driving operation reception unit 100, and the vehicle information acquisition unit 22. Based on the traveling road speed information included in the road information, the steady traveling speed of the traveling road is calculated as an upper limit speed that satisfies the speed constraint condition regarding the steady traveling represented by the following equation (11). Specifically, the steady travel restriction unit 112 is based on the position [x _t−1 , y _t−1 ] of the vehicle at time t−1 and is obtained from the road speed information of the position, and the vehicle is currently traveling. A parameter p _load (steady travel speed) representing the general maximum speed of the _road is acquired and set as the upper limit speed. The _load is calculated from past travel data, speed limit, etc. on the road.

The follow-up restriction unit 114 includes [v _t-1 ], which is an element of the state vector X _{t-1 at} the time t−1 received by the driving operation reception unit 100, and the forward vehicle information acquired by the vehicle information acquisition unit 22 ( The upper limit speed that satisfies the speed constraint condition regarding the forward vehicle following shown in the following equation (12) is calculated based on the relative speed with respect to the forward vehicle. Specifically, the follow-up restriction unit 114 determines the inter-vehicle distance with the preceding vehicle T seconds ahead by constant speed prediction based on the vehicle speed [v _t-1 ] at the time t−1 and the relative speed between the preceding vehicle and the preceding vehicle. A predicted distance ^ x _f (t + T) is calculated. Then, the tracking restriction unit 114 calculates an upper limit speed that satisfies the restriction condition of Expression (12).

Here, _pf is a parameter obtained in advance, and L is the length of the vehicle.

The curve restriction unit 116 is acquired by the vehicle information acquisition unit 22 and [x _t−1 , y _t−1 ] that are elements of the state vector X _{t−1 at} the time t−1 received by the driving operation reception unit 100. Based on the road curvature information included in the map information, the upper limit speed that satisfies the speed constraint condition relating to the curve travel shown in the following equation (13) is calculated. Specifically, the curve restriction unit 116 is based on the position [x _t−1 , y _t−1 ] of the vehicle at time t−1, and the predicted value of the absolute value of the curvature of the runway T seconds ahead at the position. | ^ Δ (t + T) | is calculated, and an upper limit speed satisfying Expression (13) is calculated based on the calculated predicted value | ^ δ (t + T) |.

Note that _pc is a parameter obtained in advance.

Target speed selection unit 118 among the upper limit speed calculated by the target speed constraint unit 108, selects the lowest rate as the target speed v _r.

The speed control unit 120 is an element of the target speed v _r selected by the target speed selection unit 118 and the state vector X _t−1 of the time t−1 received by the driving operation reception unit 100 [v _t−1 ]. Based on the above, the amount of operation of the accelerator pedal is predicted. Then, the speed control unit 120 calculates the acceleration α of the vehicle based on the predicted accelerator pedal operation amount. For example, the speed control unit 120 uses a proportional control model to predict the accelerator pedal operation amount according to the difference between the target speed v _r and the estimated vehicle speed v _{t−1 at} time t−1. Then, the acceleration α is calculated.

The driving operation output unit 122 calculates [α _t−1 , δ _t−1 ], which are elements of the state vector X _{t−1 at} the time t−1 received by the driving operation receiving unit 100, and the lateral position control unit 106. Based on the calculated curvature δ and the acceleration α calculated by the speed controller 120, the curvature δ _t and the acceleration α _{t at} time t are predicted. For example, the driving operation output unit 122 predicts the curvature δ _t at time t so as to suppress the difference between the curvature δ _{t−1 at} time t−1 and the calculated curvature δ. Similarly, the driving operation output unit 122 predicts the acceleration α _t at time t so as to suppress the difference between the acceleration α _{t-1 at} time t-1 and the calculated acceleration α. Further, the driving operation output unit 122 outputs the acceleration α _t and the curvature δ _t .

The vehicle behavior prediction unit 28 is based on the curvature δ _t and acceleration α _t output by the driving operation output unit 122 and the state vector X _{t-1 at} time t−1 estimated by the driving operation vehicle state estimation unit 32. Thus, the position information and motion state of the vehicle at time t are predicted. As shown in FIG. 6, the vehicle behavior prediction unit 28 includes a vehicle behavior reception unit 280, a speed calculation unit 282, a direction calculation unit 284, a position calculation unit 286, and a vehicle behavior output unit 288.

The vehicle behavior accepting unit 280 accepts the state vector X _t-1 at time t−1 estimated by the driving operation vehicle state estimating unit 32.

The speed calculation unit 282 is an element of the vehicle acceleration α _t output by the driving operation output unit 122 and the state vector X _t−1 of the time t−1 received by the vehicle behavior receiving unit 280 [v _t−1. ], The vehicle speed v _t is predicted. For example, the speed calculation unit 282 predicts the speed v _t of the vehicle according to the speed calculation formula v _t = v _t−1 + α _t T.

The direction calculation unit 284 includes the curvature δ _t output by the driving operation output unit 122 and [θ _t-1 ] that is an element of the state vector X _{t-1 at} the time t−1 received by the vehicle behavior receiving unit 280. Based on the vehicle speed v _t predicted by the speed calculation unit 282, the vehicle orientation θ _t is predicted.

The position calculation unit 286 includes the vehicle speed v _t predicted by the speed calculation unit 282, the vehicle direction θ _t predicted by the direction calculation unit 284, and the state at the time t−1 received by the vehicle behavior reception unit 280. The position [x _t , y _t ] of the vehicle is predicted based on [x _t−1 , y _t−1 ] which is an element of the vector X _t−1 .

The vehicle behavior output unit 288 includes the vehicle speed v _t predicted by the speed calculation unit 282, the vehicle direction θ _t predicted by the direction calculation unit 284, and the vehicle position predicted by the position calculation unit 286 [x _{t 1} , y _t ] are output as vehicle position information and motion state.

In addition, the vehicle behavior prediction unit 28 is based on a state vector X _{t−1 that} includes the vehicle position information, the motion state, and the driving operation amount of the driver at time t−1 estimated by the driving operation vehicle state estimation unit 32. The prior error covariance matrix P _{t | t−1} is calculated according to the above equations (5) to (7).

  The position information acquisition unit 30 sequentially acquires vehicle position information measured by the position measurement unit 18.

The driving operation vehicle state estimation unit 32 is based on the position information of the vehicle at time t acquired by the position information acquisition unit 30 and the probability distribution of the position information of the vehicle at time t output by the vehicle behavior output unit 288. The probability distribution of the motion state of the vehicle at time t and the probability distribution of the driving operation amount of the driver of the vehicle are estimated.
Specifically, the driving operation vehicle state estimation unit 32 outputs the vehicle position information [x ^* _t , y ^* _t ] at time t acquired by the position information acquisition unit 30 and the vehicle behavior output unit 288. Based on the position information [x _t , y _t ] of the vehicle at time t and the prior error covariance matrix P _{t | t−1} calculated by the vehicle behavior prediction unit 28, the above equations (11) to (12) Thus, the motion state of the vehicle and the driving operation amount of the vehicle driver (state vector X _{t | t} ) are estimated. Further, the driving operation vehicle state estimation unit 32, based on the prior error covariance matrix P _{t | t−1} calculated by the vehicle behavior prediction unit 28, according to the above equation (13), the posterior error covariance matrix P _{t | t} is calculated.

The output device 40 outputs the state vector X _t = [x _t y _t θ _t v _t α _t δ _t ] output by the prediction unit 24.

<Operation of Vehicle Behavior Prediction Device 10>
Next, operation | movement of the vehicle behavior prediction apparatus 10 which concerns on 1st Embodiment is demonstrated. In the following description, the driving operation amount of the driver at the next time (time t), the position information of the vehicle, and the motion state of the vehicle are predicted, and then the driving operation amount prediction process of the driver in the driving operation prediction unit 26 is performed. And the vehicle position information and the vehicle motion state prediction process in the vehicle behavior prediction unit 28 are further repeated S-1 times (S is, for example, 30), and the vehicle position information and the vehicle motion state after 3 seconds. Is to be predicted. In addition, the state vector predicted in the repeated processing up to 3 seconds later is expressed as X _{t + s} = [x _{t + s} , y _{t + s} , θ _{t + s} , v _{t + s} , α _{t + s} , δ _{t + s} ].

  First, when destination information is input by the input operation unit 12 when the vehicle travels by the driving operation of the driver of the prediction target vehicle and the position information of the vehicle is sequentially measured by the position measurement unit 18, the computer 20 In FIG. 7, the vehicle behavior prediction processing routine shown in FIG. 7 is executed.

  In step S <b> 100, the destination information input by the input operation unit 12 is received by the vehicle information acquisition unit 22.

  In step S <b> 102, the position information acquisition unit 30 receives the vehicle position information measured by the position measurement unit 18.

In step S104, based on the vehicle position information received in step S102, the state vector X _t , the posterior error covariance matrix P _{t | t} , the variance Q _{t of} the system noise u _t (error term), and the observation noise w The initial state (t = 0) of the variance R _t of _t (error term) is set, and 0 is substituted for t. In step S105, the time t is incremented by one.

In step S106, the driving operation receiving unit 100, is set in the step S104, or accepts the state vector _{X t-1} at time t-1 which is estimated in the previous step S114.

In step S108, the destination information acquired in step S100 by the prediction unit 24, the position information of the vehicle at time t-1 received in step S106, the motion state of the vehicle, and the driving operation amount of the driver of the vehicle ( Based on the state vector X _t-1 ), the driving operation amount of the driver at time t, vehicle position information, and the vehicle motion state (state vector X _t ) are predicted. Step 108 is realized by the prediction processing routine shown in FIG.

<Prediction processing routine>
First, in step S200, 0 is substituted for the number of repetitions s.

Next, in step S202, the driving operation prediction unit 26 determines the state vector X _{t-1 at} time t-1 received in step S106 or the driving operation of the driver predicted in the previous step S202 and the previous step S204. The driving operation of the driver is predicted based on the amount, the vehicle position information, and the vehicle motion state _{Xt + s-1} . Step 202 is realized by the driving operation prediction processing routine shown in FIG.

<Driving operation prediction processing routine>
The driving operation prediction processing routine, when the repeat count s in the prediction processing routine is zero, the time t-1 of the state vector _{X t-1} the following steps S302~ step S316 based on the received in step S104 Each process is performed. On the other hand, when the number of repetitions s in the prediction processing routine is not 0, the driving operation amount of the driver, the vehicle position information, and the vehicle motion state predicted in the previous step S202 and the previous step S204 in the prediction processing routine. The following steps S302 to S316 are performed based on Xt _{+ s-1} .

  First, in step S300, the vehicle information acquisition unit 22 acquires the external environment information of the vehicle at time t-1.

In step S302, the destination information received in step S100, the external environment information received in step S300, and the state vector X _t-1 of time t-1 received in step S104 are received by the route generation unit 102. Based on the element [x _t−1 , y _t−1 ] or [x _{t + s−1} , y _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204, the destination Generate the shortest route to the ground.

In step S304, the target position selection unit 104 generates the path information generated in step S302 and the element of the state vector X _t-1 at time t-1 received in step S104 [x _t−1 , y _t−1 ] or [x _{t + s−1} , y _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204, and selects the target position [x _r , y _r ] on the route. To do.

In step S306, the target position [x _r , y _r ] selected in step S304 by the lateral position control unit 106 and the elements of the state vector X _t-1 at time t−1 received in step S104. [X _t−1 , y _t−1 , v _t−1 , θ _t−1 ] or [x _{t + s−1} , y _{t + s−1} , v _{t + s} predicted by the vehicle behavior prediction unit 28 in the previous step _S <b> 204. ₋₁ , θ _{t + s−1} ], the driver's steering operation amount is predicted. Then, the curvature δ of the vehicle is calculated based on the predicted steering operation amount.

In step S308, an upper limit speed that satisfies each of the plurality of speed constraint conditions is calculated.
In this step, first, the forward / backward position control restriction unit 110 uses [x _t−1 , y _t−1 ], which is an element of the state vector X _t−1 at time t−1 received in step S104, or the previous time. The upper limit speed is calculated based on [x _{t + s−1} , y _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in step S204 and the target position [x _r , y _r ] selected in step S304. To do.
Next, [x _t−1 , y _t−1 ] which is an element of the state vector X _{t−1 at} the time t−1 received in the above step S104 by the steady travel restriction unit 112 or the vehicle in the previous step S204. Based on [x _{t + s−1} , y _{t + s−1} ] predicted by the behavior prediction unit 28 and the road speed information included in the external environment information received in step S300, the steady road speed of the road is _expressed by the above equation. (V) is calculated as the upper limit speed of the speed constraint condition shown in (8). [V _t-1 ], which is an element of the state vector X _{t-1 at} the time t-1 received by the tracking constraint unit 114 in step S104. or predicted by the vehicle behavior prediction unit 28 in the previous step S204 and _{[v t + s-1]} , the front included in the external environment information received in step S300 Based on the information (the relative speed to the preceding vehicle), and calculates the maximum speed of the speed constraints shown in equation (9).
Then, [x _t−1 , y _t−1 ], which is an element of the state vector X _t−1 at time t−1 received in step S104, or the vehicle behavior prediction in the previous step S204 by the curve restriction unit 116. Based on [x _{t + s−1} , y _{t + s−1} ] predicted by the unit 28 and the road curvature information included in the map information of the external environment information received in step S300, the speed constraint _expressed by the above equation (10). The upper limit speed of the condition is calculated.

In step S310, the by the target speed selector 118, among the upper limit speed calculated in step S308, selects the lowest rate as the target speed _{v r.}

In step S312, the the speed control unit 120, and the target speed _{v r} that is selected in step S310, the a state vector _{X t-1} element of the time t-1 that has been received in step S104 _{[v t-1]} Alternatively, the accelerator pedal operation amount is predicted based on [v _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204. Then, the vehicle acceleration α is calculated based on the predicted accelerator pedal operation amount.

In step S314, [α _t−1 , δ _t−1 ], which is an element of the state vector X _t−1 at time t−1 received in step S104, by the driving operation output unit 122, or in the previous step S202. Based on [α _{t + s−1} , δ _{t + s−1} ] predicted by the driving operation prediction unit 26, the curvature δ calculated in step S 306, and the acceleration α calculated in step S 312, at time t + s. The curvature δ _{t + s} and the acceleration α _{t + s} (when s = 0, the curvature δ _t and the acceleration α _t ) are predicted.

In step S316, the curvature δ _{t + s} and the acceleration α _{t + s} and the acceleration α _{t + s} (when s = 0, the curvature δ _t and the acceleration α _t ) predicted in step S314 are output as the center value of the driving operation. The driving operation prediction processing routine is terminated.

Next, returning to the prediction processing routine, in step S204, the vehicle behavior prediction unit 28 predicts the state vector X _{t-1 at} time t-1 received in step S106, or the previous step S202 and the previous step S204. Based on the driving operation amount of the driver, the vehicle position information, and the vehicle motion state X _{t + s−1} , the vehicle position information and the vehicle motion state at time t (or time t + s) are predicted. Step 204 is realized by the vehicle prediction processing routine shown in FIG.

<Vehicle prediction processing routine>
The predicted vehicle processing routine, when the repeat count s in the prediction processing routine is zero, the time t-1 that has been received in step S104 state vectors _{X t-1} Based on the following steps S400~ step S404 When each processing is performed and the number of repetitions s in the prediction processing routine is not 0, the driving operation amount of the driver, the vehicle position information, and the vehicle predicted in the previous step S202 and the previous step S204 in the prediction processing routine Each of the following steps S400 to S404 is performed based on the exercise state Xt _{+ s-1} .

First, in step S400, the velocity calculation unit 282 outputs the vehicle acceleration α _t output in step S202 and the element of the state vector X _t−1 at time t−1 received in step S106 [v _{t −1} ] or [v _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204, the vehicle speed v _{t + s} (v _t when s = 0) is predicted.

Next, in step S402, the orientation calculation unit 284, and a curvature [delta] _t which is output in step S202, an element of the state vector _{X t-1} at time t-1 that has been received in the step S106 [theta _{t- 1} ] or [θ _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204 and the vehicle speed v _t predicted in the step S400, the vehicle direction θ _{t + s} is predicted. To do.

In step S404, the vehicle speed v _{t + s} predicted in step S400, the vehicle direction θ _{t + s} predicted in step S402, and the time t−1 received in step S106 are detected by the position calculation unit 286. [X _t−1 , y _t−1 ] which is an element of the state vector X _t−1 of the above, or [x _{t + s−1} , y _{t + s−1} ] predicted by the vehicle behavior prediction unit 28 in the previous step S204. Based on the above, the position [x _{t + s} , y _{t + s} ] of the vehicle is predicted.

In step S406, the vehicle speed v _{t + s} of the vehicle at time t predicted in step S400, the vehicle orientation θ _{t + s at} time t predicted in step S400, and the vehicle speed at time t predicted in step S404. The position [x _{t + s} , y _{t + s} ] is output as the vehicle behavior center value, and the vehicle prediction processing routine is terminated.

  Next, returning to the prediction processing routine, in step S206, it is determined whether or not the number of repetitions s is equal to or greater than S-1. If the number of repetitions s is greater than or equal to S-1, the process proceeds to step S208. On the other hand, if the number of repetitions s is less than S-1, s is incremented by 1 in step S210, and the process returns to step S202.

In step S208, the vehicle behavior predicting unit 28 sets the state vector X _{t-1 including} the vehicle position information, the motion state, and the driving operation amount of the driver at time t-1 estimated by the driving operation vehicle state estimating unit 32. Based on the above equations (5) to (7), the prior error covariance matrix P _{t | t−1} is calculated.

In step S210, the curvature δ _{t + s} and the acceleration α _{t + s} (s = 0,..., S−1) at time t to t + S−1 output in step S202, and the time t to output from step S204. The vehicle position [x _{t + s} , y _{t + s} ], the vehicle orientation θ _{t + s} , and the vehicle speed v _{t + s} (s = 0,..., S−1) at _{t +} S−1, and the advance calculated in step S208 above. The error covariance matrix P _{t | t−1} is output as a result, and the prediction processing routine is terminated.

  Next, returning to the vehicle behavior prediction processing routine, in step S110, the output device 40 outputs the result output in step S108 as a prediction result from time t to 3 seconds later.

  In step S112, the position information acquisition unit 30 receives the position information of the vehicle at time t.

In step S114, the position information [x ^* _t , y ^* _t ] of the vehicle at time t received in step S112 and the vehicle position at time t output in step S110 are received by the driving operation vehicle state estimation unit 32. Based on the information [x _t , y _t ], the motion state of the vehicle and the driving operation amount of the driver of the vehicle are estimated. Step 114 is realized by the driving state vehicle state estimation processing routine shown in FIG.

In step S500, the curvature δ _t , acceleration α _t , vehicle speed v _t , vehicle direction θ _t , and vehicle position [x _t , y _t ] at time t output in step S110 are received, The state vector at time t is configured as X _t = [x _t y _t θ _t v _t α _t δ _t ].

In Step S502, the prior error covariance matrix P _{t | t−1} output in Step S110, the observation matrix H _t, and the variance R of the observation noise w _t (error term) are output by the driving vehicle state estimation unit 32. Based on _t , the Kalman gain K _t is calculated according to the above equation (11).

In step S504, the Kalman gain K _t calculated in step S502, the state vector X _{t at} time t configured in step S500, and the vehicle position received in step S112 are calculated by the driving vehicle state estimation unit 32. Based on Y _t = [x ^* _t , y ^* _t ] and the observation matrix H _t , the state vector ^ X _t = [^ x _t , ^ y _t , ^ θ at time t according to the above equation (12) _t , ^ v _t , ^ α _t , ^ δ _t ] are estimated.

In step S506, the prior error covariance matrix P _{t | t−1} calculated in step S206, the Kalman gain K _t calculated in step S502, and the observation matrix H _t are calculated by the driving vehicle state estimation unit 32. Then, a posterior error covariance matrix P _{t | t} is calculated.

In step S508, the state vector at time t, which is estimated in step _{S504 ^ X t = [^ x} t, ^ y t, ^ θ t, ^ v t, ^ α t, ^ δ t] and, in step S506 As a result, the a posteriori error covariance matrix P _{t | t} calculated in the above is output, and the driving operation vehicle state estimation processing routine is terminated.

  Next, the process returns to the vehicle behavior prediction process routine, and the process returns to step S105 to increment the time t by 1.

  As described above, according to the vehicle behavior prediction apparatus according to the first embodiment, based on the acquired position information of the vehicle at time t and the predicted position information of the vehicle at time t. The vehicle motion state at t and the driving operation amount of the vehicle driver are estimated, and based on the external environment information of the vehicle at time t, and the estimated vehicle motion state and driving operation amount of the vehicle driver at time t. , The driving operation amount of the driver of the vehicle at time t + 1 is predicted, and the position information of the vehicle at time t + 1 is based on the predicted driving operation amount of the driver at time t + 1 and the estimated motion state of the vehicle at time t + 1. And by predicting the motion state, it is possible to accurately predict the vehicle position information and the motion state based on the external environment information of the vehicle and the vehicle position information.

  Further, since it is possible to predict the vehicle position information and the motion state from only the external environment information and the position information, it is possible to obtain a prediction performance with higher accuracy than in the past.

  Further, since the driving operation amount of the driver can be predicted from the position information and the external environment information, a vehicle not equipped with the device (for example, a vehicle equipped with the vehicle behavior prediction device according to the present embodiment). The behavior of the vehicle ahead) can be predicted.

  In addition, since it is possible to predict the vehicle position information and the motion state from only the external environment information and the position information, it is possible to perform the prediction process at a higher speed than in the past.

  Further, since it is possible to predict the vehicle position information and the motion state from only the external environment information and the position information, the vehicle position information and the motion state are predicted without using sensor information such as the vehicle speed and acceleration. be able to.

  In addition, an upper limit speed that satisfies each of a plurality of predetermined speed constraint conditions is calculated, one of the calculated upper limit speeds is selected as a target speed, and the selected target speed and the vehicle at time t are selected. Based on the external environment information, the map information, and the estimated motion state of the vehicle at time t, the driving operation amount of the driver of the vehicle at time t + 1 is predicted, so that the position information and motion state of the vehicle can be accurately obtained. Can be predicted.

  In addition, since a plurality of driving scenes are assumed, the prediction accuracy of general road driving is high.

  In addition, compared with the prior art, it has high speed prediction accuracy, can predict behavior only with external environment information and a position sensor such as GPS, and is highly robust to observation errors of the position sensor.

  In addition, a driver model for steady driving, forward vehicle tracking, signal stop, and curves, which occupies most of the driving on general roads, is introduced as a speed constraint condition, and it is a model that performs speed control for the target speed that satisfies those speed constraint conditions. By predicting driving operations and predicting vehicle behavior, it is possible to estimate the probability distribution of driving operations and vehicle states using the extended Kalman filter from the observed position information, and prediction of behavior is possible only with external environment information and position sensors such as GPS. It becomes possible, and both accuracy and robustness are improved.

<Second Embodiment>
Next, a second embodiment will be described. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  The second embodiment is different from the first embodiment in that the driving operation amount of the driver is predicted by further considering the signal information, and the position information and the motion state of the vehicle are predicted.

  As shown in FIG. 12, the vehicle behavior prediction apparatus 210 according to the second embodiment includes an input operation unit 12 for a driver to input a destination, map information (road network data), road information, and a signal. A road network database 214 storing information, a forward vehicle information acquisition unit 16 that detects a forward vehicle traveling ahead of the prediction target vehicle, a position measurement unit 18 that measures the position of the prediction target vehicle, and a prediction of the behavior of the vehicle And a computer 20 that outputs the prediction result.

  The road network database 214 stores map information, road information, and signal information. Here, the signal information includes data indicating the position of the traffic signal and the signal change cycle of the traffic signal.

  The vehicle information acquisition unit 222 further acquires signal information on the running path of the vehicle from the road network database 214 as external environment information of the vehicle, and acquires a signal state at each time t based on the acquired signal information. To do.

  As shown in FIG. 13, the driving operation prediction unit 226 includes a driving operation reception unit 100, a route generation unit 102, a target position restriction unit 2101, a target position selection unit 2104, a lateral position control unit 106, and a target speed. A restriction unit 108, a target speed selection unit 118, a speed control unit 120, and a driving operation output unit 122 are provided.

The target position restriction unit 2101 is based on the external environment information of the vehicle at time t−1 acquired by the vehicle information acquisition unit 22 and the state vector X _{t−1 at} time t−1 received by the driving operation reception unit 100. Thus, a vehicle position that satisfies a plurality of position constraint conditions is calculated. The target position restriction unit 2101 includes a signal restriction unit 2102 and a route restriction unit 2103.

The signal restriction unit 2102 includes the signal information included in the external environment information of the vehicle at time t−1 acquired by the vehicle information acquisition unit 22 and the state vector X _{t− at} time t−1 received by the driving operation reception unit 100. _{Based on} [x _t−1 , y _t−1 , v _t−1 ] which is an element of ₁ , the vehicle position that satisfies a predetermined position constraint condition regarding signal stop is calculated. Specifically, the signal restriction unit 2102 calculates a vehicle position [x _r , y _r ] that satisfies the position restriction condition represented by the following expression (14).

Here, [x _r , y _r ] ^T is the vehicle position, [x _t−1 , y _t−1 ] ^T is the position at time t−1 received by the driving operation reception unit 100, and [x _stopline , y _stopline ] ^T is the position of the stop line corresponding to the signal, S _t-1 is the state of the previous signal (red, yellow, blue), S _red is the red signal, and v _t-1 is the time received by the driving operation receiving unit 100 The speed of the vehicle at t−1, p _signal, is a parameter determined in advance for determining whether to stop.

The route restriction unit 2103 is an element of the route information obtained based on the destination of the vehicle by the route generation unit 102 and the state vector X _{t−1 at} the time t−1 received by the driving operation reception unit 100 [x. _{Based on [t−1} , y _t−1 ]], the vehicle position that satisfies the position constraint condition regarding the route is calculated. For example, the route restriction unit 2103 calculates 100 points ahead from the position [x _t−1 , y _t−1 ] at the time t−1 on the route to the destination as the vehicle position.

The target position selection unit 2104 selects either the vehicle position calculated by the signal restriction unit 2102 or the vehicle position calculated by the route restriction unit 2103, and determines the selected position as the target position. For example, the target position selection unit 2104 selects the vehicle position closest to the position [x _t−1 , y _t−1 ] at time t−1 as the target position.

  In addition, about the other structure and effect | action of the vehicle behavior prediction apparatus which concern on 2nd Embodiment, since it is the same as that of 1st Embodiment, description is abbreviate | omitted.

  As described above, according to the vehicle behavior prediction apparatus according to the second embodiment, a vehicle position that satisfies a predetermined position constraint condition regarding signal stop is calculated based on the signal information, and the destination of the vehicle is calculated. By calculating a vehicle position that satisfies a position constraint condition regarding the route based on the route information obtained based on the route, selecting one of the calculated vehicle positions, and determining the selected position as a target position The position information and the motion state of the vehicle can be predicted with high accuracy.

In the above embodiment, the case where the own vehicle on which the vehicle behavior prediction device is mounted is set as a prediction target vehicle has been described as an example. However, the present invention is not limited to this, and the vehicle behavior prediction device is mounted. A vehicle other than the host vehicle that is not used may be used as the prediction target vehicle.
When a vehicle not equipped with a vehicle behavior prediction device is used as a prediction target vehicle, the vehicle-mounted device of the prediction target vehicle includes an input operation unit 12, a forward vehicle information acquisition unit 16, and a position measurement unit 18. The prediction device acquires external environment information and position information of the prediction target vehicle through communication means, and predicts the position information and motion state of the vehicle.

  Moreover, although the case where the vehicle behavior prediction apparatus is mounted on the vehicle has been described as an example, the present invention is not limited to this, and even if the vehicle behavior prediction apparatus is realized by a server set in a traffic center or the like. Good.

  In the above embodiment, map information, road information (road speed information and road curvature information), forward vehicle information, and signal information are used as external environment information of the vehicle. However, the present invention is not limited to this. In order to deal with more various scenes, information such as oncoming vehicles, surrounding vehicles, and pedestrians may be used.

  Moreover, although the front vehicle information acquisition part 16 in said embodiment demonstrated the case where a laser radar apparatus was used as an example, it is not limited to this. For example, the front vehicle information acquisition unit 16 may acquire the front vehicle information using a vehicle-to-vehicle communication device, a road-to-vehicle communication device, or the like.

  Further, in the above-described embodiment, the case where the above-described vehicle behavior prediction apparatus includes the road network database 14 is described as an example. However, the present invention is not limited to this. For example, the road network database 14 is a vehicle. The vehicle behavior prediction device provided in the external device of the behavior prediction device may refer to the road network database 14 by communicating with the external device using communication means.

  In the prediction processing routine in the above embodiment, the case where the driving operation prediction process and the vehicle behavior prediction process are repeated 30 times has been described as an example. However, the present invention is not limited to this. It is not necessary to repeat the behavior prediction process.

  In the above embodiment, the target speed constraint unit 108 calculates an upper limit speed that satisfies each of the plurality of speed constraint conditions, and the target speed selection unit 118 selects the lowest one of the upper limit speeds as the target speed. However, the present invention is not limited to this, and the target speed may be selected based on each combination of a plurality of speed constraint conditions.

Further, the driving operation output unit 122 in the above embodiment is calculated by [α _t−1 , δ _t−1 ], which is an element of the state vector X _t−1 at time t−1, and the lateral position control unit 106. The case where the curvature δ _t and the acceleration α _{t at the} time t are predicted based on the calculated curvature δ and the acceleration α calculated by the speed control unit 120 has been described as an example, but the present invention is not limited to this. Without using [α _t−1 , δ _t−1 ], which is an element of the state vector X _t−1 at time t−1, the curvature δ _t and acceleration α _{t at} time t can be predicted. Good.

  Moreover, although the driving operation vehicle state estimation part 32 in said embodiment demonstrated the case where the movement state of the vehicle in time t and the driving operation amount of the driver of a vehicle were estimated as an example, it is not limited to this. Alternatively, only the motion state of the vehicle at time t may be estimated.

  The program of the present invention can be provided by being stored in a recording medium.

DESCRIPTION OF SYMBOLS 10,210 Vehicle behavior prediction apparatus 12 Input operation part 14,214 Road network database 16 Forward vehicle information acquisition part 18 Position measurement part 20 Computer 22, 222 Vehicle information acquisition part 24 Prediction part 26,226 Driving operation prediction part 28 Vehicle behavior prediction Unit 30 position information acquisition unit 32 driving operation vehicle state estimation unit 40 output device 100 driving operation reception unit 102 route generation unit 104, 2104 target position selection unit 106 lateral position control unit 108 target speed restriction unit 110 front and rear position control restriction unit 112 steady state Travel restriction unit 114 Follow restriction unit 116 Curve restriction unit 118 Target speed selection unit 120 Speed control unit 122 Driving operation output unit 280 Vehicle behavior reception unit 282 Speed calculation unit 284 Direction calculation unit 286 Position calculation unit 288 Vehicle behavior output unit 2101 Target position Constraint unit 2102 Signal constraint unit 2103 Constraints section

Claims (7)

Acquisition means for acquiring vehicle position information and external environment information of the vehicle at each time t; external environment information of the vehicle at time t-1 acquired by the acquisition means; and information on the vehicle at time t-1. Driving operation prediction means for predicting the driving operation amount of the driver of the vehicle at time t based on the exercise state;
Based on the driving operation amount of the driver at time t predicted by the driving operation prediction means and the motion state of the vehicle at time t-1, the position information of the vehicle and the motion state of the vehicle at time t are obtained. Vehicle behavior prediction means for predicting;
Driving that estimates the motion state of the vehicle at time t based on the position information of the vehicle at time t acquired by the acquisition means and the position information of the vehicle at time t predicted by the vehicle behavior prediction means. Operating vehicle state estimating means;
Including
The driving operation prediction means is based on the external environment information of the vehicle at time t acquired by the acquisition means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. Predicting the driving operation amount of the driver of the vehicle,
The vehicle behavior prediction means is based on the driving operation amount of the driver at time t + 1 predicted by the driving operation prediction means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. A vehicle behavior prediction apparatus that predicts position information of the vehicle and a motion state of the vehicle at time t + 1. The driving operation prediction means is based on the external environment information of the vehicle at time t acquired by the acquisition means and the probability distribution of the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. , Predicting the probability distribution of the driving amount of the driver of the vehicle at time t + 1,
The vehicle behavior prediction means includes a probability distribution of the driving operation amount of the driver at time t + 1 predicted by the driving operation prediction means, and a motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. Based on the probability distribution, the probability distribution of the position information of the vehicle at time t + 1 and the probability distribution of the motion state of the vehicle are predicted,
The driving operation vehicle state estimation means is based on the position information of the vehicle at time t + 1 acquired by the acquisition means and the probability distribution of the position information of the vehicle at time t + 1 predicted by the vehicle behavior prediction means. The vehicle behavior prediction apparatus according to claim 1, wherein a probability distribution of the motion state of the vehicle at time t + 1 is estimated. Target speed selection means for calculating an upper limit speed satisfying each of a plurality of predetermined speed constraint conditions based on the external environment information, and selecting any one of the calculated upper limit speeds as a target speed. Including
The driving operation predicting means is estimated by the target speed selected by the target speed selecting means, the external environment information of the vehicle at time t acquired by the acquiring means, and the driving operation vehicle state estimating means. The vehicle behavior prediction apparatus according to claim 1 or 2, wherein a driving operation amount of a driver of the vehicle at the time t + 1 is predicted based on a motion state of the vehicle at the time t. For each time t, further includes target position determining means for determining a target position on the running path of the vehicle,
The acquisition means includes, as external environment information of the vehicle, a steady vehicle traveling speed on the road on which the vehicle travels, a curvature of the road on which the vehicle travels, and a distance from the vehicle traveling in front of the vehicle. Information for each time t,
The target speed selection means calculates the steady travel speed as an upper limit speed that satisfies a speed constraint condition regarding steady travel, calculates an upper limit speed that satisfies a speed constraint condition regarding curve travel based on the curvature of the road, Based on the preceding vehicle information, an upper limit speed that satisfies a speed constraint condition regarding forward vehicle tracking is calculated, and an upper limit condition that satisfies the speed constraint condition regarding front-rear position control is calculated based on the target position on the road determined by the target position determination unit. The vehicle behavior prediction apparatus according to claim 3, wherein a speed is calculated, and any one of the calculated upper limit speeds is selected as a target speed. The acquisition means further acquires signal information on the running path of the vehicle and a destination of the vehicle for each time t as external environment information of the vehicle,
The target position determining means calculates a vehicle position that satisfies a predetermined position constraint condition related to signal stop based on the signal information, and based on the route information obtained based on the destination of the vehicle, The vehicle behavior prediction device according to claim 4, wherein a vehicle position that satisfies a position constraint condition for the vehicle is calculated, any one of the calculated vehicle positions is selected, and the selected position is determined as a target position. The driving operation prediction means includes the external environment information of the vehicle at time t acquired by the acquisition means, the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means, and the driver of the vehicle. Based on the driving operation amount, the driving operation amount of the driver of the vehicle at time t + 1 is predicted,
The driving operation vehicle state estimation means is based on the position information of the vehicle at time t + 1 acquired by the acquisition means and the position information of the vehicle at time t + 1 predicted by the vehicle behavior prediction means. The vehicle behavior prediction apparatus according to claim 1, wherein the movement state of the vehicle and a driving operation amount of a driver of the vehicle are estimated. Computer
Acquisition means for acquiring vehicle position information and external environment information of the vehicle at each time t; external environment information of the vehicle at time t-1 acquired by the acquisition means; and information on the vehicle at time t-1. Driving operation prediction means for predicting the driving operation amount of the driver of the vehicle at time t based on the exercise state;
Based on the driving operation amount of the driver at time t predicted by the driving operation prediction means and the motion state of the vehicle at time t-1, the position information of the vehicle and the motion state of the vehicle at time t are obtained. Based on the vehicle behavior prediction means for prediction, the vehicle position information at time t acquired by the acquisition means, and the vehicle position information at time t predicted by the vehicle behavior prediction means, the vehicle at time t A program for functioning as a driving vehicle state estimating means for estimating a motion state of a vehicle,
The driving operation prediction means is based on the external environment information of the vehicle at time t acquired by the acquisition means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. Predicting the driving operation amount of the driver of the vehicle,
The vehicle behavior prediction means is based on the driving operation amount of the driver at time t + 1 predicted by the driving operation prediction means and the motion state of the vehicle at time t estimated by the driving operation vehicle state estimation means. A program for predicting the position information of the vehicle and the motion state of the vehicle at time t + 1.