VEHICLE DRIVING CONTROL DEVICE AND VEHICLE DRIVING CONTROL METHOD

Abstract

A vehicle travel control device (10) for controlling the travel of an own vehicle based on a predicted route which is a future travel route of the own vehicle, the device comprising: Object detection means (11, 12) for detecting the position of a stationary object located on the roadway or along the side of the roadway on which the host vehicle is traveling; Position storage means (25) for time-serially storing a position of a preceding vehicle, which is a position of the preceding vehicle traveling ahead of the host vehicle; Movement trajectory determining means (20) for comparing a movement locus of another vehicle, which is the movement trajectory of the position of the preceding vehicle stored in the position storing means, with the position of the fixed stationary object detected by the object detecting means to determine whether or not the movement locus of the other vehicle lies along the roadway; and Course calculation means (20) for validating the movement locus of the other vehicle when the movement locus of the other vehicle has been determined by the movement trajectory determination means to be along the road course, for invalidating the movement locus of the other vehicle when it is determined that it is not along the road course, and for calculating the predicted route based on the validated movement locus of the other vehicle.

Description

[Technical field]

[0001] The present invention relates to a vehicle travel control system, and more particularly, to a travel control technology for controlling travel of an own vehicle based on a predicted route of the own vehicle. [State of the art]

[0002] Vehicle following control, in which a host vehicle follows a preceding vehicle traveling in the same lane as the host vehicle, alongside preceding vehicles traveling ahead of the host vehicle, is known as an example of vehicle driving support control. It is important that such vehicle following control correctly selects the vehicle traveling in the same lane as the host vehicle from among the preceding vehicles detected by a sensor, a camera, or the like. Therefore, it is conventional to calculate a future travel route of the host vehicle and set the preceding vehicle located on the future travel route as the target for the vehicle following control. Furthermore, various methods for calculating the future travel route of the host vehicle have been proposed (see, for example, JP 2002-531886 A).JP 2002 – 531 886 A describes that a travel locus of the preceding vehicle, which is traveling in front of the host vehicle, is stored and the stored travel locus is used to calculate the future travel route of the host vehicle.

[0003] The above-mentioned JP 2002-531886 A does not consider the case where a vehicle deviates from the lane, such as the case where the preceding vehicle performs an unexpected lane change. Therefore, when such a situation occurs, there is a risk that the calculation accuracy of the route prediction for the subject vehicle will decrease. Furthermore, the movement locus of the preceding vehicle can only be obtained through an inter-vehicle section between the subject vehicle and the preceding vehicle, so there is a problem that the prediction of the future route is within a limited range.

[0004] Furthermore, a method for course prediction in driver assistance systems for motor vehicles is known from DE 10 2005 002 719 A, in which, in a case in which an obstacle, such as a construction site barrier, is present in front of a vehicle, the course of a lane object is validated or made valid when the vehicle in front crosses the lane marking.

[0005] From DE 101 15 909 A1 a method for selecting a preceding vehicle is known, in which an experimentally determined map is used to determine the probability that the preceding vehicle is driving in the same lane as the own vehicle, depending on a deviation of a preceding vehicle from a center of the own vehicle.

[0006] An object of the present invention is to provide a vehicle travel control technology capable of increasing the prediction accuracy of the travel route of an own vehicle. [Solution to the task]

[0007] The present invention uses the following measures.

[0008] The present invention relates to a vehicle travel control device for controlling the travel of an own vehicle based on the predicted route, which is the future travel route of the own vehicle. The travel control device of the present invention includes object detection means for detecting the position of a stationary object on the road or along the road on the side of the road on which the own vehicle is traveling; position storage means for time-serially storing the position of the preceding vehicle, which is the position of the preceding vehicle traveling ahead of the own vehicle, a movement locus of another vehicle, which is the movement trajectory of the position of the preceding vehicle, stored in the position storage means;Trajectory determining means for determining whether or not the moving locus of the other vehicle is along the roadway by comparing it with the position of a stationary object detected by the object detecting means; and course calculating means for validating the moving locus of the other vehicle when the trajectory determining means determines that the moving locus of the other vehicle is along the roadway, invalidating the moving locus of the other vehicle when it determines that the moving locus of the other vehicle is not along the roadway, and calculating the predicted route based on the confirmed moving locus of the other vehicle.

[0009] When using the movement locus of the preceding vehicle to perform route prediction of the host vehicle, there is a problem that the calculation accuracy in the route prediction for the host vehicle is reduced when using data when the preceding vehicle makes an unexpected lane change, etc., or makes a movement that does not conform to the shape of the road. In view of this point, the travel control device of the present invention is configured to determine whether or not the movement locus of the preceding vehicle lies along the shape of the road by comparing the intersection lines of the traveling road and the position of a stationary object, such as a guardrail or the like, on the road side with the movement locus of the preceding vehicle.Furthermore, if it is determined that the movement locus of the preceding vehicle follows the road pattern, the movement locus of the preceding vehicle is validated. However, if it is determined that the movement locus of the preceding vehicle does not follow the road pattern, the movement locus of the preceding vehicle is invalidated. Furthermore, the travel control device of the present invention is configured to calculate the predicted route of the vehicle based on the validated movement locus of the preceding vehicle. The travel control device of the present invention invalidates the movement locus of the preceding vehicle when the preceding vehicle unexpectedly makes a movement that does not conform to the road pattern, and can thus suppress the predicted route for the vehicle from being erroneously calculated.

[0010] Furthermore, the travel control device of the present invention is a vehicle travel control device for controlling the travel of the host vehicle based on the predicted route which is the future travel route of the host vehicle, and includes a position storage means for time-serially storing the position of the preceding vehicle which is of the preceding vehicle traveling ahead of the host vehicle;Radius estimation means for estimating the curve radius of the roadway to be traveled by the host vehicle, and course calculation means for calculating the predicted route for the inter-vehicle section from the host vehicle to the preceding vehicle based on the movement locus of the other vehicle, which is the movement trajectory of the position of the preceding vehicle, stored in the position storage means, and for calculating the predicted route for a section farther away than the inter-vehicle section in which the predicted route calculated based on the movement locus of the other vehicle is extended based on the curve radius estimated by the radius estimation means;

[0011] The movement locus of the preceding vehicle is determined only from the inter-vehicle section from the subject vehicle to the preceding vehicle, and thus the desired course based on the movement locus is limited to the inter-vehicle section. In this case, driving control decreases in controllability by restricting the vehicle to the target vehicle for driving control based on the route prediction. Taking this into account, the driving control device of the present invention expands the predicted route using the estimated curve radius, so that a travel route farther from the subject vehicle can be predicted, and a course of the subject vehicle toward the other vehicle farther ahead can be predicted.The travel control apparatus of the present invention can enlarge the selected area of the vehicle, which is the travel control target, based on the route prediction and is capable of performing travel control based on the route prediction thus made. [Short description of the drawing] Fig. 1 is a block diagram showing the schematic structure of a vehicle driving control device. Fig. 2 is a diagram showing the locus of motion of the preceding vehicle when traveling on a curved road. Fig. Figure 3 is an explanatory diagram of the case where a dividing line and a guardrail are stationary objects. Fig. Figure 4 is a diagram explaining the processes involved in calculating a locus of motion. Fig. Figure 5 is an explanatory diagram to explain the operations used to estimate R to extend the predicted route. Fig. Figure 6 is a flowchart showing the process flows of a predicted route calculation process. Fig. Figure 7 is a flowchart showing the process flows of the predicted route calculation process based on the movement locus. [Description of the embodiments]

[0012] An embodiment of a vehicle travel control device will be explained below with reference to the drawing. The travel control device according to the present embodiment is installed in a vehicle and performs vehicle following control for traveling behind a preceding vehicle, which is selected from preceding vehicles traveling in front of the own vehicle and traveling in the same lane as the own vehicle. The vehicle following control controls the distance between the own vehicle and the preceding vehicle. First, using Fig. 1 explains the schematic structure of the travel control device of the present embodiment.

[0013] In Fig. 1, the travel control device 10 is a computer including a central processing unit (CPU), a random access memory (RAM), a read-only memory (ROM), an input / output (I / O), etc. The travel control device 10 includes a route prediction unit 20, a following vehicle setting unit 35, and a control target calculation unit 36, and the CPU realizes each of these functions by executing a program installed in the ROM. An object detection device for detecting an object located near the vehicle is arranged in the vehicle (own vehicle). The travel control device 10 receives the detection information of the object from the object detection device and performs vehicle following control with respect to the preceding vehicle based on the input information. An imaging device 11 and a radar device 12 are arranged in the vehicle as the object detection device.

[0014] The imaging device 11 is an on-board camera and is configured by a charge-coupled device (CCD) camera, a CMOS (complementary metal-oxide-semiconductor) image sensor, a near-infrared camera, or the like. The imaging device 11 captures images of the surroundings including the roadway traveled by the host vehicle and generates image data indicative of the captured image, which is sequentially output to the travel control device 10. The imaging device 11 is installed, for example, near the upper side of the front windshield of the host vehicle and captures images of a region extending over a certain angular range θ1 in the forward direction of the vehicle with respect to the center of the imaging axis. The imaging device 11 may be a single-lens camera or a stereo camera.

[0015] The radar device 12 is a detection device for detecting objects by transmitting electromagnetic waves as transmission waves and receiving the reflected waves. The present embodiment is constituted by a millimeter-wave radar. The radar device 12 is mounted on the front of the host vehicle, and a radar signal scans the area extending over a certain angular range θ2 (θ2 < θ1) in the forward direction of the vehicle with respect to the center of the optical axis. Furthermore, the radar device 12 forms distance measurement data based on the time until the reflected waves are received after the electromagnetic waves are transmitted in the forward direction of the vehicle, and sequentially outputs the generated distance measurement data to the travel control device 10. The distance measurement data includes information regarding the direction in which the object exists, the distance to the object, and the relative speed.

[0016] It should be noted that the imaging device 11 and the radar device 12 are each mounted such that the imaging axis, which is the reference axis of the imaging device 11, and the optical axis, which is the reference axis of the radar device 12, are in the same direction, corresponding to a direction parallel to the traveling road surface of the own vehicle. The detectable range of the imaging device 11 and the detectable range of the radar device 12 overlap each other at least in part. It should be noted that the radar device 12 corresponds to a "first detection device," and the imaging device 11 to a "second detection device."

[0017] The driving control device 10 receives the image data from the imaging device 11 and the distance measurement data from the radar device 12, and receives corresponding detection signals from each sensor in the vehicle. A yaw rate sensor 13 for detecting the angular velocity (hereinafter referred to as "yaw rate") around the vertical axis of the vehicle, a speed sensor 14 for detecting the speed, etc., are provided as the various sensors. Furthermore, a steering angle sensor 15 for detecting the steering angle and an ACC switch 16 that is operated when the driver selects the vehicle following control mode are provided.

[0018] The route prediction unit 20 is an arithmetic unit for predicting the travel route of the host vehicle and is equipped with a first predicted route arithmetic unit 21 and a second predicted route arithmetic unit 22. Of these units, the first predicted route arithmetic unit 21 calculates the future travel route of the host vehicle based on the movement locus of the preceding vehicle traveling ahead of the host vehicle. Furthermore, the second predicted route arithmetic unit 22 calculates the future travel route of the host vehicle based on the yaw rate of the host vehicle.

[0019] The first predicted route calculation unit 21 receives fixed object information from a fixed object information acquisition unit 23, white line information from a white line information acquisition unit 24, and the other vehicle's moving locus information from an other vehicle's moving locus acquisition unit 25. The first predicted route calculation unit 21 calculates the first predicted route, which is the future travel route of the own vehicle, by combining the input information. Note that the first predicted route calculation unit 21 is capable of route prediction for the own vehicle, which is independent of the own vehicle's yaw rate.

[0020] The stationary object information acquisition unit 23 calculates the position information regarding stationary road-side objects (three-dimensional objects such as guardrails, walls, or the like) located along the road on which the host vehicle is traveling based on the distance measurement data from the radar device 12 and outputs the position information as stationary object information to the first predicted route arithmetic unit 21. The white line information acquisition unit 24 calculates information regarding road dividing lines (white lines) included in the image from the imaging device 11 based on the image data from the imaging device 11 and outputs the calculated information to the first predicted route arithmetic unit 21 as white line information.Regarding the concrete calculation method for the white line information, for example, the white line information acquisition unit 24 extracts the corner points considered as candidates for a white line from the image data based on the change rate, etc., of the brightness in the horizontal direction of the image. Furthermore, the white line information acquisition unit 24 sequentially stores the extracted corner points in one frame and calculates the white line information based on the stored history of the white line corner points.

[0021] The other vehicle movement locus acquisition unit 25 calculates the position of the preceding vehicle, which is the position of the preceding vehicles (coordinates corresponding to the passing points of the preceding vehicles), in a certain cycle based on the distance measurement data (the distance information between the own vehicle and the preceding vehicle and the position information in the lateral direction) from the radar device 12, and stores the calculated position of the preceding vehicle in a certain storage area (storage device) on a time-serial basis.The other-vehicle movement locus acquisition unit 25 calculates the movement locus of the preceding vehicle based on the time-series data of the stored position of the preceding vehicle and outputs the calculated movement locus as movement locus information for the other vehicle to the first predicted route calculation unit 21. Thus, the other-vehicle movement locus acquisition unit 25 functions as a position storage device. Note that the other-vehicle movement locus acquisition unit 25 calculates the movement locus information not only for vehicles traveling in the same lane as the own vehicle among the preceding vehicles, but also for vehicles traveling in a lane adjacent to the own vehicle, and this calculation is used for route prediction of the own vehicle.

[0022] The second predicted route calculation unit 22 receives the estimated turning radius (hereinafter referred to as "estimated R") of the roadway traveled by the host vehicle from a turning radius estimation unit 26, and the input estimated R is used to calculate the second predicted route, which is the future travel route of the host vehicle. The turning radius estimation unit 26 calculates the estimated R from the yaw rate (yaw angle) from the yaw rate sensor 13 and the speed detected by the speed sensor 14. The calculation method for the estimated R is not limited to this, and the estimated R can be calculated, for example, using image data or can be calculated using the steering angle from the steering angle sensor 15 and the speed detected by the speed sensor 14. The turning radius estimation unit 26 thus functions as a radius estimation device.

[0023] A following vehicle setting unit 35 uses the predicted route calculated by the route prediction unit 20 and sets the vehicle located on the predicted route as the following vehicle from among the preceding vehicles traveling ahead of the host vehicle. The control target value calculation unit 36 calculates the control target value for maintaining the distance between the following vehicle set by the following vehicle setting unit and the host vehicle by controlling the traveling speed of the host vehicle. Note that in this case, the control target value calculation unit 36 calculates the control target value for maintaining the distance between the vehicles at a set target interval.Specifically, control values such as the target output of the own vehicle's engine and the requested braking energy are calculated, and these control values are output to the engine electronic control unit (engine ECU 41) as a control signal. Note that, in the present embodiment, the travel control device 10 is configured to output the control signal to the engine ECU 41, and the control signal from the engine ECU 41 to the brake electronic control unit (brake ECU 42). Regarding the output configuration of the control signal, the travel control device 10 can output a control signal to each of the engine ECU 41 and the brake ECU 42.

[0024] Regarding the route prediction of the own vehicle, the present embodiment confirms the route prediction result based on the route prediction result calculated by the first predicted route arithmetic unit 21, that is, based on the movement locus of the preceding vehicle, and uses the result to select the vehicle to follow. The reason for this is as follows. When traveling on a straight road, the prediction accuracy hardly changes for the first predicted route, which is the route prediction result based on the movement locus of the preceding vehicle, and the second predicted route, which is the route prediction result based on the estimated R.

[0025] However, when the vehicle to be followed enters a curve and the host vehicle is still traveling on a straight road before reaching the curve, if the second predicted route is used to select the vehicle to be followed, there is a risk that the preceding vehicle located in an adjacent lane will be selected as the vehicle to be followed, rather than the preceding vehicle in the same lane as the host vehicle. Therefore, the present embodiment preferably uses the first predicted route to select the vehicle to be followed.

[0026] When using the movement locus of the preceding vehicle to perform route prediction of the host vehicle, if data is used for a preceding vehicle that makes a movement that does not correspond to the shape of the road, for example, when the preceding vehicle makes an unexpected lane change, there is a risk that the calculation accuracy in the route prediction for the host vehicle will decrease. Taking this into account, the travel control device 10 according to the present embodiment uses the detection result of the stationary objects, such as the dividing lines of the traveling road and stationary road-side objects (a three-dimensional object such as a guardrail, a wall, or the like) along the road, and compares the position of a stationary object with the movement locus of the preceding vehicle.If the result determines that the movement locus of the preceding vehicle follows the course of the road, the movement locus of the preceding vehicle is declared valid. However, if the movement locus of the preceding vehicle is determined not to follow the course of the road, the movement locus of the preceding vehicle is declared invalid. Furthermore, the first predicted route is calculated based on the valid movement locus of the preceding vehicle.

[0027] Note that the configuration for not confirming the movement locus of the preceding vehicle includes the configuration for deleting the movement locus calculated using the position of the preceding vehicle and the configuration for prohibiting the use of the movement locus calculated using the position of the preceding vehicle. Furthermore, the configuration for deleting the position of the preceding vehicle or the vehicle detection point by the radar device 12 and the configuration for prohibiting their use are also included.

[0028] The operations for calculating the movement locus of the preceding vehicle, the operations for calculating the first predicted route based on the movement locus, and the operations for calculating the second predicted route based on the estimated R are explained below. Fig. 2 is a diagram showing the movement locus RT of the preceding vehicle M2 in the case where the host vehicle M1 and the preceding vehicle M2 are traveling on a curved road.

[0029] Fig. 2 shows a plurality of leading vehicle positions Pd resulting from the detection of the leading vehicle M2 by the radar device 12, and a plurality of stationary object detection points Pa resulting from the detection of a three-dimensional object (e.g., a road-side guardrail) by the radar device 12 as a stationary road-side object. Note that the leading vehicle position Pd averages the plurality of vehicle detection points that are the detection results by the radar device 12 at a certain distance interval (e.g., 10 m). Fig. Figure 2(a) shows the case where the motion locus RT of the preceding vehicle M2 conforms to (follows) the roadway, and (b) shows the case where the motion locus RT does not conform to (follow) the roadway. In the following explanation, the motion locus RT of the preceding vehicle M2 conforming to the roadway is referred to as "along the roadway," and the motion locus RT of the preceding vehicle M2 that does not conform to the roadway is referred to as "not along the roadway."

[0030] Fig. Figure 2(a) shows the state where the preceding vehicle M2 does not change lanes and continuously travels in the same lane as the host vehicle, so the movement locus RT calculated from the time-series data of the preceding vehicle position Pd is along the road. In this case, when the movement locus RT of the preceding vehicle M2 (each of the preceding vehicle positions Pd) is compared with the plurality of stationary object detection points Pa of the three-dimensional object (the stationary object), it is found that the relative positions in the lateral direction relative to the traveling direction of the host vehicle M1 are substantially constant. Therefore, the movement locus RT of the preceding vehicle M2 follows the road during traveling and is thus validly recognized as the movement locus used for route prediction of the host vehicle M1.

[0031] In contrast, Fig. 2(b) shows the state in which the preceding vehicle M2 changes lanes to the right lane, and the movement locus RT calculated from the time-series data of the positions Pd of the preceding vehicle is no longer along the roadway. In this case, when the movement locus RT of the preceding vehicle M2 (each position Pd of the preceding vehicle) is compared with the plurality of fixed object detection points Pa of the three-dimensional object (fixed object), it is found that the relative positions in the lateral direction are not constant with respect to the traveling direction of the host vehicle M1. Thus, the movement locus RT of the preceding vehicle M2 is not along the roadway during traveling and is thus recognized as an invalid movement locus for use in the route prediction of the host vehicle 1. It should be noted that in Fig. 2 the position of a stationary object, which is located on the left in the direction of travel of the own vehicle M1, is compared with the locus of movement RT of the preceding vehicle M2, but a stationary object which is located on the right in the direction of travel of the own vehicle M1 can also be used as a comparison target.

[0032] Fig. 3 is an explanatory view of a case where dividing lines (white lines) and the guardrail are set as stationary objects for the host vehicle M1, and each of these positions is calculated. Fig. Figure 3 shows fixed object detection points Pa, which are recognized as guardrails on the road side, and the dividing lines (white lines) Pb for dividing the lanes on the road. Furthermore, Fig. 3(a) a parking bay area J (area where the vehicle can leave the carriageway) located in a part of the side of the carriageway in the straight line of the road. Fig. However, Figure 3(b) shows a roadside construction fence. In this case, the location of the fixed object detection point Pa does not coincide with the dividing line Pb.

[0033] If in Fig. 3(a) When the preceding vehicle M2 moves straight, the movement locus RT of the preceding vehicle M2 can be obtained as linear. Furthermore, the dividing line Pb is also detected as linear. On the other hand, the fixed object detection point Pa corresponding to the exit J is detected as a lateral extension. When the movement locus RT of the preceding vehicle M2 (each of the positions Pd of the preceding vehicle) is compared with the plurality of fixed object detection points Pa of the three-dimensional object (stationary object), it is found that the relative positions in the lateral direction with respect to the traveling direction of the host vehicle M1 are constant. On the other hand, when the movement locus RT of the preceding vehicle M2 is compared with the dividing line Pb, it is found that the relative positions in the lateral direction with respect to the traveling direction of the host vehicle M1 are constant.In this case, the movement locus RT of the preceding vehicle M2 is recognized as a valid movement locus used in the route prediction of the own vehicle M1 based on the relative positions of the movement locus RT and the dividing line Pb, which are constant in the lateral direction.

[0034] Furthermore, Fig. 3(b), the preceding vehicle M2 intentionally moves along the guide fence rather than along the line Pb. Therefore, when the movement locus RT of the preceding vehicle M2 (each position Pb of the preceding vehicle) is compared with the plurality of fixed object detection points Pa of the three-dimensional object (fixed object), it is found that the relative positions in the lateral direction with respect to the traveling direction of the host vehicle M1 are constant. On the other hand, when the movement locus RT of the preceding vehicle M2 is compared with the dividing line Pb, it is found that the relative positions in the lateral direction with respect to the traveling direction of the host vehicle M1 are not constant.In this case, the movement locus RT of the preceding vehicle M2 is recognized as valid for the movement locus used in the route prediction of the own vehicle M1 based on the relative positions of the movement locus RT and the fixed object detection point Pa, which are constant in the lateral direction.

[0035] The following are based on Fig. 4 explains the procedures for calculating the movement locus RT to use the time-series data of the position Pd of the preceding vehicle used for the movement locus RT of the preceding vehicle M2, which is recognized as confirmation of the movement locus used for the route prediction of the own vehicle M1 to specify the vehicle to be followed.

[0036] In the present embodiment, according to Fig. 4 a plurality of sections K (K1 - K5 in Fig. 4) at intervals of, for example, 10 m in the forward direction of the own vehicle M1 and the positions Pd for the preceding vehicle (Pd1 - Pd5 in Fig. 4) are calculated for each section. Furthermore, three or more sections are defined as a unit section KN and the straight carriageway α (α1 - α3 in Fig. 4) is calculated based on the position Pd of the preceding vehicle for the unit sections KN. In this case, in the present embodiment, the unit sections KN are set to partially overlap each other in the inter-vehicle section from the host vehicle M1 to the preceding vehicle M2. Furthermore, the movement locus RT for specifying the vehicle to be followed is calculated by forming the straight road α for each unit section KN.

[0037] Specifically, in the first embodiment, first, the sections K1-K3 are set as a unit section KN (1-3) to calculate the straight road α1. Subsequently, the sections K2-K4 are set as a unit section KN (2-4) to calculate the straight road α2. Subsequently, the sections K3-K5 are set as a unit section KN (3-5) to calculate the straight road α3. Here, when calculating the straight road α1 of the unit sections KN (1-3), the straight road α1 is calculated by connecting the positions Pd1 and Pd3 of the preceding vehicle, which correspond to the section K1 and the section K3, respectively, and which are located on both sides of and sandwich the section K2, with a straight line. The other unit sections are calculated in the same way.It should be noted that the method for calculating the straight path α of the unit section KN is not limited to the above method of connecting the leading vehicle positions Pd (two leading vehicle positions) of the respective sections K located on both sides of the unit section KN by means of a straight line. For example, it is possible to calculate all the leading vehicle positions Pd (in . Fig. 4 three such vehicle positions) contained in each unit section KN to calculate the straight roadway α by a linear approximation process or the like. Furthermore, the unit section KN may contain four or more sections K.

[0038] Furthermore, in the present embodiment, regarding the calculated straight lane α1-α3, each straight lane α1-α3 is connected by an appropriate shift in the lateral direction. Here, in connecting (binding) each straight lane α1-α3, the start point (the end point on the subject vehicle M1 side) of the straight lane α1 is set as the forward position of the subject vehicle M1. Subsequently, the start point of the straight lane α2 is set as the position Pd2 of the preceding vehicle on the straight lane α1, and the start point of the straight lane α3 is set as the position Pd4 of the preceding vehicle on the straight lane α2. In the present embodiment, the positions Pd1, Pd2, Pd4, and Pd5 of the preceding vehicle are connected with a straight line and thus connected (composed) to each straight lane α1-α3.In the present embodiment, by calculating the movement locus RT in this way, even if noise (spatial noise) is included in one of the positions Pd for the preceding vehicle, it is possible to remove the noise to calculate the movement locus RT for determining the vehicle to be followed.

[0039] Note that there are cases where a plurality of leading vehicles M2 exist in the forward direction of the host vehicle M1, and the positions of these leading vehicles M2 are in front and behind each other (for example, when the leading vehicles M2 exist in the same lane as the host vehicle M1 in the adjacent lane). In this case, the final motion locus R3 can be calculated by calculating the motion loci RT for the respective leading vehicles M2 by the above method and connecting each of the calculated RTs.The connection of the movement loci RT can be achieved, for example, by shifting the movement locus RT of the preceding vehicle M2 away from (at the rear of) the subject vehicle M1 in the lateral direction relative to the movement locus RT of the preceding vehicle M2 close to (in front of) the subject vehicle M1 between the two preceding vehicles M2. In this case, the length of the movement locus RT can extend further than in the case where the movement locus RT is calculated based on the position Pd of the preceding vehicle by a preceding vehicle M2.

[0040] Furthermore, for example, when two leading vehicles M2 are present, if the motion locus RT of one of the leading vehicles M2 and the motion locus RT of the other leading vehicle are discontinuous, only the motion locus RT of the leading vehicle M2 near the host vehicle M1 can be confirmed. Alternatively, the motion locus RT can be interpolated by the estimated value R, which is the estimated curve radius of the road, to estimate the discontinuous section between the two motion loci RT by using the turning information (the yaw rate and the steering angle) and the speed of the host vehicle M1, or by estimating from the estimated R estimated from the image data.

[0041] The history (the time-series data of the leading vehicle positions Pd) of the leading vehicle M2 can only be obtained from the inter-vehicle section between the host vehicle M1 and the leading vehicle M2. Therefore, the predicted route of the host vehicle M1, which can be calculated based on the movement locus RT of the leading vehicle M2, is limited to this inter-vehicle section. Therefore, in the present embodiment, the first predicted route is calculated based on the movement locus RT of the leading vehicle M2 as described above for the inter-vehicle section between the host vehicle M1 and the leading vehicle M2.Furthermore, the predicted route of the host vehicle M1 is calculated by extending the first predicted route calculated based on the movement locus RT of the preceding vehicle M2 by the second predicted route calculated based on the estimated R, which is the estimated curvature radius of the roadway for the section farther away than the inter-vehicle section. Therefore, in the present embodiment, it is possible to predict the travel route of the host vehicle M1 for a longer distance ahead.

[0042] Fig. 5 is an explanatory view for explaining a process that uses the first predicted route based on the movement locus RT of the preceding vehicle M2 and the second predicted route based on the estimated R, which is the estimated curve radius of the road, to extend the predicted route of the host vehicle M1. Fig. 5, the first predicted route RA based on the movement locus RT of the preceding vehicle M2 is shown by a dotted line, and the second predicted route RB based on the estimated R is shown by a dash-double-dotted line, and the predicted route RC of the host vehicle M1 composed and extended therefrom is shown by a solid line.

[0043] In Fig. 5, in the area ahead of the host vehicle M1, the area S1 near the host vehicle M1 is the area where the history of the positions of the preceding vehicle M2 (time-series data of the positions Pd of the preceding vehicle) can be obtained (exists). However, the area S2 farther from the host vehicle M1 than the area S1 is the area where the history of the positions of the preceding vehicle M2 cannot be obtained (exists). The present embodiment calculates the movement locus RT of the preceding vehicle M2 from the time-series data of the position Pd of the preceding vehicle for the inter-vehicle section (area S1) between the host vehicle M1 and the preceding vehicle M2. Subsequently, the first predicted route RA is calculated based on the calculated movement locus RT.Furthermore, the second predicted route RB is calculated based on the estimated R for the section (area S2) that is farther away than the inter-vehicle section (area S1). In the present embodiment, the second predicted route RB calculated based on the estimated R is connected to the position PE of the first predicted route RA, which is the end position of the movement locus RT of the preceding vehicle M2. The predicted route is extended, and the extended predicted route RC of the host vehicle M1 is thereby calculated.

[0044] Hereinafter, the calculation process of the predicted route of the travel control device 10 according to the present embodiment will be described based on the Fig. 6 and Fig. 7. These processes are performed by the route prediction unit 20. Furthermore, these processes are performed at every specified period by the ECU of the travel control device 10 during vehicle travel and when the ACC switch 16 is turned on.

[0045] As in Fig. As shown in Fig. 6, the traveling control device 10 acquires the time-series data (the history of the positions of the preceding vehicle M2) of the position Pd of the preceding vehicle in step S101. When there are a plurality of preceding vehicles M2, the determination processes of the following step S103 and step S104 are performed for the movement locus RT of the vehicles. Thereafter, in the following step S102, the traveling control device 10 determines the position of a stationary object (stationary object detection point Pa). In step S102, a three-dimensional object (e.g., a guardrail, a median strip, a guide fence, etc.) on the roadway or along the road side and the dividing lines Pb on the roadway are set as stationary objects, and their position information is obtained.The position information of the three-dimensional object is obtained by the distance measurement data from the radar device 12 and the position information of the dividing line Pb is obtained from the image data from the imaging device 11.

[0046] Subsequently, in the driving control device 10, the movement locus RT of the preceding vehicle M2 calculated in step S3 based on the time-series data of the position Pd of the preceding vehicle is compared with the position of the three-dimensional object, and it is determined whether or not the movement locus RT lies along the three-dimensional object (whether or not it lies along the roadway). Step S103 determines whether or not the relative positions of the position Pd of the preceding vehicle and the fixed object detection point Pa in the lateral direction relative to the traveling direction of the host vehicle M1 coincide by comparing the plurality of positions Pd of the preceding vehicle from step S101 with the plurality of fixed object detection points Pa of the three-dimensional object from step S102.Specifically, the travel control device 10 extracts the combination of the position Pd of the preceding vehicle and the fixed object detection point Pa where the position in the traveling direction of the host vehicle M1 is the same, and calculates the lateral distance between the position Pd of the preceding vehicle and the fixed object detection point Pa at a plurality of points. When the difference in the distance compared at the plurality of points is a certain value or less, it is determined that the relative positions in the lateral direction with respect to the traveling direction of the host vehicle M1 are consistent. Thus, in the travel control device, the route prediction unit 20 functions as a trajectory determining device.

[0047] It should be noted that the method for determining whether or not the movement locus RT of the preceding vehicle M2 is along the three-dimensional object (whether or not it is along the roadway) is not limited to the above method. For example, the determination can be made based on a distance between any point on the movement locus RT where the position is the same as the traveling direction of the host vehicle M1 and any point on the line connecting the fixed object detection points Pa.

[0048] Further, in step S104, the traveling control device 10 compares the movement locus RT calculated based on the time-series data of the position Pd of the preceding vehicle with the dividing line Pb, and determines whether or not the movement locus RT is along the dividing line Pb (whether or not it is along the roadway). Step S104 determines whether or not the relative positions of the movement locus RT and the dividing line Pb in the lateral direction with respect to the traveling direction of the host vehicle M1 coincide. More specifically, the traveling control device 10 extracts a plurality of relative positions in the lateral direction between the movement locus RT relative to the traveling direction of the host vehicle M1 and the dividing line Pb.When the difference between the relative positions in the comparison of the plurality of points is a certain value or less, it is determined that the relative positions in the lateral direction with respect to the traveling direction of the own vehicle M1 agree.

[0049] The movement trajectory generated by connecting the plurality of positions Pd of the preceding vehicle by a straight line is used as the movement locus RT of the preceding vehicle M2, which is used in the comparison process of steps S103 and S104 to easily perform the determination of valid / invalid of the movement locus RT. With a view to increasing the determination accuracy, the movement locus Rt of the preceding vehicle M2 obtained by the above calculation method according to Fig. 4 can be used in the valid / invalid determination process of the motion locus RT.

[0050] As a result, if the driving control device 10 determines in step S103 that the movement locus Rt of the preceding vehicle M2 is along the three-dimensional object (along the roadway) (YES in S103), the flow proceeds to step S105. Similarly, if the driving control device 10 determines in step S104 that the movement locus RT of the preceding vehicle is along the dividing line Pb (along the roadway) (YES in S104), the flow proceeds to step S105. The driving control device 10 recognizes (judges) in step S105 that the movement locus RT of the preceding vehicle M2 is valid. Subsequently, the driving control device 10 calculates the predicted route (first predicted route RA) based on the movement locus RT determined to be confirmed in step S106. Therefore, in the travel control device 10, the route prediction unit 20 operates as a course calculation device.

[0051] However, if the driving control device 10 determines in step S103 that the movement locus RT of the preceding vehicle M2 is not along the three-dimensional object (is not along the roadway) (NO in S103), the flow proceeds to step S104. Further, if the driving control device 10 determines in step S104 that the movement locus RT of the preceding vehicle is not along the dividing line Pb (is not along the roadway) (NO in S104), the flow proceeds to step S107. The driving control device 10 recognizes (judgments) in step S107 that the movement locus RT concerning the preceding vehicle M2 is not valid. If there is no valid movement locus RT, the predicted route (second predicted route RB) is calculated based on the estimated R, and the predicted route is used to set the vehicle to follow.

[0052] The following is based on the use of Fig. 7, the calculation process for the predicted route (process of step S106) based on the movement locus RT is explained. As in Fig.As shown in Fig. 7, the travel control device 10 calculates the straight lane α in the unit sections KN based on the time-series data of the position Pd of the preceding vehicle (step S201). Subsequently, in step S202, the movement locus RT is calculated by composing the straight lanes α (connection in the straight lane α) of the unit section KN, and the movement locus RT is used to calculate the predicted route (first predicted route RA) of the inter-vehicle distance in step S203. When a plurality of preceding vehicles M2 exist in the forward direction of the host vehicle M1 and the movement locus RT of the plurality of preceding vehicles M2 is determined to be confirmed, the plurality of movement loci RT determined to be confirmed are composed.Furthermore, when there is a preceding vehicle M2 for which the movement locus RT is confirmed, the predicted route (first predicted route RA) of the inter-vehicle section is calculated by taking a weighted average of the white line information concerning the movement locus RT of the preceding vehicle M2. However, when there are a plurality of preceding vehicles M2 for which the movement locus RT is confirmed, the predicted route (first predicted route RA) of the inter-vehicle section is calculated by taking a weighted average with the white line information concerning the movement locus RT composed thereof.

[0053] Subsequently, the travel control device 10 extends the predicted route of the own vehicle M1 into the section (area S2) farther than the inter-vehicle section (area S1) based on the estimated R in step S204. More specifically, the second predicted route RB calculated based on the estimated R is connected to the first predicted route RA calculated based on the moving locus RT to extend the predicted route, so that the extended predicted route RC of the own vehicle M1 is obtained. After that, the travel control device 10 ends this process. The present embodiment uses the predicted route calculated by this process to perform the specification of the vehicle to be followed.

[0054] The above present embodiment can obtain the following excellent results.

[0055] The driving control device 10 according to the present embodiment can determine whether or not the movement locus RT is along the road course by comparing the position of a stationary object (stationary object detection point Pa), such as the dividing line Pb of the traveling lane and the guardrail, etc., on the side of the road, with the movement locus RT of the preceding vehicle M2. Further, if it is determined that the movement locus RT is along the road course, the movement locus RT of the preceding vehicle is validated. On the other hand, if it is determined that the movement locus RT is not along the road course, the movement locus RT is invalidated. Further, the predicted route of the host vehicle M1 is calculated based on the valid movement locus RT.Therefore, the travel control device 10 according to the present embodiment invalidates the movement locus RT of the preceding vehicle M2 when the preceding vehicle M2 makes an unexpected movement that does not correspond to the roadway, so that the predicted route of the host vehicle M1 can be suppressed from being erroneously calculated.

[0056] The driving control device 10 according to the present embodiment can validate the movement locus RT when a three-dimensional object located on the road or along the road side and a dividing line Pb on the road are set as stationary objects, and the movement locus RT of the preceding vehicle M2 is located along at least one of the three-dimensional objects and the dividing line Pb. A variety of objects other than guardrails, such as roadside trees and indicator boards, can be assumed as a three-dimensional object on the road side, and for estimating the road course, there are many objects that may be a source of interference. Therefore, when the three-dimensional object is excluded from the comparison target with the movement locus RT of the preceding vehicle M2, only a comparison is performed between the dividing line Pb and the movement locus RT.As a result, there is a risk that the determination accuracy will decrease due to the narrow range within which the dividing line Pb is recognizable. Taking this into account, the driving control device 10 according to the present embodiment maintains a balance between suppressing the decrease in prediction accuracy due to inappropriate use of data and ensuring the ability to execute the predicted route based on the movement locus RT through the above configuration, and thus, driving control can be performed based on the route prediction.

[0057] The movement locus RT of the preceding vehicle M2 can be obtained only for the inter-vehicle section between the host vehicle M1 and the preceding vehicle M2. Thus, the predicted route of the host vehicle M1, which can be calculated based on the movement locus RT of the preceding vehicle M2, is limited to the inter-vehicle section. Therefore, the travel control device 10 according to the present embodiment is configured to use the estimated value R, which is the estimated curve radius of the road, to extend the predicted route of the host vehicle M1. As a result, the travel control device 10 according to the present embodiment can predict the travel route further ahead of the host vehicle M1 and can predict the course of the host vehicle M1 to a further-away preceding vehicle M2.The selected area of the vehicle used as the target for the driving control based on the route prediction can thus be enlarged and it is possible to perform the driving control based on the route prediction.

[0058] The travel control device 10 according to the present embodiment is configured to calculate the straight road α in the unit sections KN, calculate the movement locus RT of the preceding vehicle M2 based on the composition of the straight vehicle α, and use the calculated movement locus RT to calculate the predicted route (first predicted route RA) for vehicle following control. The travel control device 10 according to the present embodiment can eliminate spatial noise, and thus, route prediction can be performed with high accuracy even when the traveling road has a complicated shape, such as an S-curve. (Other embodiments)

[0059] The present invention is not limited to the above embodiment and can be carried out as follows.

[0060] • In the above embodiment, the first predicted route calculation unit 21 is configured to receive the stationary object information, as well as the white line information and the movement locus information of the other vehicle, and calculate the predicted route using the input information. The method for calculating the predicted route is not limited to this, and, for example, a method that does not perform weighted averaging with the white line information may be used.

[0061] • The above embodiment is configured such that, when the three-dimensional object on the road or along the road edge and the dividing line Pb on the road are set as the stationary object, it is determined that the moving locus RT of each preceding vehicle M2 is along at least one of the three-dimensional object and the dividing line Pb, and then the moving locus is validated. Further, when it is determined that the moving locus RT of the preceding vehicle M2 is not along both the three-dimensional object and the dividing line Pb, the moving locus RT is invalidated. However, the configuration for performing the valid / invalid determination of the moving locus RT is not limited to this.For example, the motion locus RT can be validated if only the three-dimensional object is set as the comparison target and it is determined that the motion locus RT lies along the three-dimensional object, and the motion locus RT can be invalidated if it is determined that the motion locus RT does not lie along the three-dimensional object. Furthermore, only the dividing line Pb can be used in the comparison target with the motion locus RT.

[0062] • The above embodiment is configured such that the straight road α is calculated in the unit sections KN, and the movement locus RT of the preceding vehicle M2 is calculated based on the composition of the straight road α, but the structure for calculating the movement locus RT is not limited to this. For example, the movement locus RT may be calculated by connecting the plurality of positions Pd of the preceding vehicle with a straight line, and the movement locus RT may be calculated by averaging the plurality of vehicle detection points included in the distance measurement data.

[0063] • The above embodiment is configured to determine whether or not the movement locus RT is along the road course by comparing the dividing line Pb currently detected by the imaging device 11 and the movement locus RT of the preceding vehicle M2. However, the configuration for determining the movement locus is not limited. For example, it may be determined whether or not the movement locus RT is along the road course by comparing the predicted route in which the dividing line Pb currently detected by the imaging device 11 has been extended a long distance and the movement locus RT.

[0064] • The above embodiment is configured so that when the movement locus RT is invalidated based on the comparison result between the movement locus RT of the preceding vehicle M2 and the position of a stationary object, and a valid movement locus RT does not exist, the predicted route (second predicted route RB) is validated based on the estimated R, but this is not limited to this. For example, when there is no valid movement locus RT, route prediction may be prohibited.

[0065] • The above embodiment is configured to determine whether or not the relative positions of the moving locus RT of the preceding vehicle M2 and the stationary object detection point Pa are constant based on the positional deviations between the lateral direction relative to the traveling direction of the own vehicle M1, but the present invention is not limited to this. When determining whether or not the relative positions in the lateral direction relative to the traveling direction of the own vehicle M1 are constant, the determination can be made, for example, by comparing the shape of the dividing line Pb on the road surface and the moving locus RT, and comparing the shape of the moving locus RT and the line connected to the stationary object detection point Pa.

[0066] • The above embodiment includes the imaging device 11 and the radar device 12 as an object detection device, but is not limited thereto. The imaging device 11 and the radar device 12 can be used, for example, in a configuration in which ultrasound is used in the transmission wave to provide a sonar for object detection. Furthermore, the technology of the present invention can be applied to a vehicle in which the imaging device 11 is not installed.

[0067] • The above embodiment was explained with reference to the case where vehicle following control is performed to follow a preceding vehicle M2 located in the same lane as the host vehicle M1. The technology of the present invention can be used in route prediction of the host vehicle M1 to avoid a collision between the host vehicle M1 and the other vehicle. The technology of the present invention can be implemented in various ways, for example, with respect to the program for executing each functional unit (each device) constituting the above travel control device 10 in a computer and a medium storing the program of the vehicle travel control method. [List of reference symbols] 10 Driving control device 11 Imaging device 12 Radar device 13 Yaw rate sensor 20 route prediction unit 21 First calculation unit for the predicted route 22 Second computing unit for the predicted route 23 Fixed Object Information Acquisition Unit 24 White Line Information Acquisition Unit 25 Acquisition unit for the locus of motion of another vehicle 26 Curve radius estimation unit 35 setting unit for the following vehicle 36 Control setpoint calculation unit 41 Engine ECU 42 Brake ECU

Claims

[1] A vehicle travel control device (10) for controlling the travel of an own vehicle based on a predicted route which is a future travel route of the own vehicle, the device comprising: Object detection means (11, 12) for detecting the position of a stationary object located on the roadway or along the side of the roadway on which the host vehicle is traveling; Position storage means (25) for time-serially storing a position of a preceding vehicle, which is a position of the preceding vehicle traveling ahead of the host vehicle; Movement trajectory determining means (20) for comparing a movement locus of another vehicle, which is the movement trajectory of the position of the preceding vehicle stored in the position storing means, with the position of the fixed stationary object detected by the object detecting means to determine whether or not the movement locus of the other vehicle lies along the roadway; and Course calculation means (20) for validating the movement locus of the other vehicle when the movement locus of the other vehicle has been determined by the movement trajectory determination means to be along the road course, for invalidating the movement locus of the other vehicle when it is determined that it is not along the road course, and for calculating the predicted route based on the validated movement locus of the other vehicle. [2] A vehicle travel control device according to claim 1, wherein the object recognition means comprise a first recognition means (12) for recognising a three-dimensional object on the roadway or along the roadway side of the roadway as the stationary object and second recognition means (11) for recognising a dividing line on the roadway as the stationary object, wherein the movement path determining means determines whether or not the movement locus of the other vehicle lies along the road course by determining whether or not the movement locus of the other vehicle lies along at least one of the three-dimensional object detected by the first detection means and the dividing line detected by the second detection means, and wherein the course calculation means validates the movement locus of the other vehicle when it is determined that the movement locus of the other vehicle is along at least one of the three-dimensional object and the dividing line. [3] A vehicle travel control apparatus according to claim 1 or 2, comprising radius estimation means (26) for estimating the curve radius of the road traveled by the host vehicle, wherein the course calculation means calculates the predicted route based on the movement locus of the other vehicle for an inter-vehicle section from the host vehicle to the preceding vehicle and calculates the predicted route by extending the predicted route calculated on the movement locus of the other vehicle based on the curve radius estimated by the radius estimation means for the section farther away than the inter-vehicle section. [4] A vehicle travel control method for controlling travel of an own vehicle based on a predicted route which is a future travel route of the own vehicle, the method comprising: a step (11, 12) of detecting the position of a stationary object on the roadway or along the roadway side of the roadway on which the subject vehicle is traveling, using a specific detection device; a storage step (25) of storing, on a time-serial basis, a position of a preceding vehicle, which is a position of the preceding vehicle traveling ahead of the own vehicle, in the determined storage device; a step (20: S103, S104) of comparing a movement locus of another vehicle, which is the movement trajectory of the position of the preceding vehicle stored in the storage device, with the position of the stationary object detected by the detection device to determine whether or not the movement locus of the other vehicle lies along the road course; and a validation step (20: S105, S106, S107) of the movement locus of the other vehicle when it is determined that the movement locus of the other vehicle is along the road course, a invalidation step of the movement locus of the other vehicle when it is determined that the movement locus of the own vehicle is not along the road course, and a calculation step of the validated predicted route based on the movement locus of the other vehicle.

Images