JP5287736B2 - Vehicle control device - Google Patents

Description

  The present invention relates to a vehicle control device that generates a travel plan for a host vehicle and performs vehicle control based on the travel plan.

  A technology has been developed in which a travel plan such as a speed pattern of a vehicle is made and automatic driving control is performed according to the traveling plan, or various driving support controls are performed in cooperation with the driving operation of the driver based on the traveling plan. Patent Document 1 discloses generating a travel plan for the host vehicle and performing automatic operation control according to the travel plan.

JP 2008-129804 A

  When a preceding vehicle exists in front of the host vehicle, the host vehicle may not be able to travel according to the travel plan depending on the traveling state of the preceding vehicle. In particular, in the case of a curve (including a left / right turn at an intersection), the degree and position of deceleration differ depending on the preceding vehicle, so it has a great influence on the travel plan of the host vehicle, and it is necessary to perform deceleration that is not in the travel plan. Occur. If the vehicle cannot travel according to the originally planned travel plan, conditions such as low fuel consumption when the travel plan is made cannot be achieved.

  Then, this invention makes it a subject to provide the vehicle control apparatus which makes the travel plan corresponding to the behavior at the time of the curve of a surrounding vehicle.

The vehicle control device according to the present invention is a vehicle control device that generates a travel plan of the host vehicle and performs vehicle control based on the travel plan, and the behavior of the surrounding vehicle during curve traveling is determined when the surrounding vehicle curves. The surrounding vehicle behavior predicting means for predicting based on the maximum lateral acceleration or / and the deceleration of the vehicle, and the travel plan for generating the travel plan of the own vehicle based on the behavior of the surrounding vehicle during the curve traveling predicted by the surrounding vehicle behavior predicting means And generating means and learning means for learning the maximum lateral acceleration or / and deceleration at the time of the curve driving of the surrounding vehicle when the surrounding vehicle travels the curve , and the surrounding vehicle behavior predicting means is the maximum learned by the learning means. Based on the lateral acceleration or / and deceleration, the behavior of the surrounding vehicle at the time of curve driving is predicted .

  In this vehicle control device, the surrounding vehicle behavior predicting means predicts the behavior of the surrounding vehicle during curve traveling based on the maximum lateral acceleration or / and deceleration during the curve traveling of the surrounding vehicle. The maximum lateral acceleration and deceleration at the time of running on the curve indicate the curve running characteristics of the vehicle, and are optimal parameters for predicting the situation (deceleration situation etc.) when the vehicle is running on the curve. In the vehicle control device, the travel plan generation unit generates a travel plan for the host vehicle based on the predicted behavior of the surrounding vehicle during the curve travel. In this way, the vehicle control device predicts the behavior of the surrounding vehicle at the time of the curve traveling based on the maximum lateral acceleration and the deceleration, and generates the traveling plan of the own vehicle according to the behavior of the surrounding vehicle, thereby generating the curve. Even if the behavior of surrounding vehicles changes, it is possible to make a travel plan corresponding to that. For example, even when the preceding vehicle decelerates toward the curve approach, by controlling the vehicle according to this travel plan, it is possible to reduce the speed before the preceding vehicle decelerates, such as deceleration that is not in the travel plan. There is no need to do it, and you can run as planned. As a result, it is possible to achieve conditions such as low fuel consumption when making a travel plan.

  Note that the curve travel of the surrounding vehicle includes a deceleration zone when the surrounding vehicle enters the curve. The curve includes turning left and right at intersections.

  In this vehicle control device, every time a surrounding vehicle travels a curve, the maximum lateral acceleration or / and deceleration during the curve traveling of the surrounding vehicle is learned by the learning means. In the vehicle control device, the surrounding vehicle behavior predicting means predicts the behavior of the surrounding vehicle during curve traveling based on the maximum lateral acceleration or / and deceleration obtained by the learning. Thus, in the vehicle control device, by learning the maximum lateral acceleration or / and deceleration at the time of curve traveling for the surrounding vehicle, the curve traveling characteristics of each surrounding vehicle (and thus the individual driver) can be obtained. It is possible to predict the behavior of the surrounding vehicle at the time of curve driving with high accuracy.

In the vehicle control apparatus of the present invention, the learning means learns the time differential value of the lateral acceleration when the surrounding vehicle travels the curve when the surrounding vehicle travels the curve, and the surrounding vehicle behavior prediction means is the learning means. It is preferable to predict the behavior of the surrounding vehicle at the time of curve driving based on the learned time differential value of the lateral acceleration .

  In this vehicle control device, every time a surrounding vehicle travels a curve, the time differential value (lateral jerk) of the maximum lateral acceleration when the surrounding vehicle travels the curve is also learned by the learning means. In the vehicle control device, the surrounding vehicle behavior prediction means takes into account the behavior of the surrounding vehicle during curve driving in consideration of the time differential value of the maximum lateral acceleration in addition to the maximum lateral acceleration or / and deceleration obtained by the learning. Predict. Thus, in the vehicle control device, by learning the time differential value of the maximum lateral acceleration at the time of curve traveling for the surrounding vehicle, it is possible to obtain a more accurate curve traveling characteristic of each surrounding vehicle. It is possible to predict the behavior at the time of curve driving with higher accuracy. The characteristic that the curve is smoothly bent is obtained by the time differential value of the maximum lateral acceleration.

  In the vehicle control device of the present invention, when the surrounding vehicle travels along a curve, the learning means learns the maximum lateral acceleration or / and deceleration during the curve traveling of the surrounding vehicle based on the road shape information of the curve. The surrounding vehicle behavior predicting means predicts the behavior of the surrounding vehicle at the time of curve driving based on the maximum lateral acceleration or / and deceleration for each road shape of the curve learned by the learning means according to the road shape of the traveling curve. It is preferable.

  In this vehicle control device, every time a surrounding vehicle travels a curve, the maximum lateral acceleration or / and deceleration for each road shape of the curve of the surrounding vehicle is learned by the learning means based on the road shape information of the curve. Keep it. In the vehicle control device, the surrounding vehicle behavior predicting means determines the curve of the surrounding vehicle based on the maximum lateral acceleration or / and deceleration for each road shape of the curve obtained by learning according to the road shape of the traveling curve. Predict driving behavior. As described above, the vehicle control device learns the maximum lateral acceleration and deceleration according to the road shape of the curve for the surrounding vehicle, thereby obtaining more accurate curve driving characteristics according to the road shape of the curve of each surrounding vehicle. Can be obtained, and the behavior of the surrounding vehicle during curve driving can be predicted with higher accuracy. The road shape of the curve includes the road shape of the road after the curve in addition to the road shape of the curve itself. In addition to road alignment information, the road shape includes information such as road width, number of lanes, presence of oncoming lanes, and the like.

  In the vehicle control device of the present invention, the travel plan of the host vehicle is a travel plan for the travel speed of the host vehicle, and the behavior of the surrounding vehicle during the curve travel is a change in the travel speed during the curve travel of the surrounding vehicle. is there.

  The present invention predicts the behavior of surrounding vehicles during curve driving based on the maximum lateral acceleration and deceleration, and generates the driving plan of the own vehicle according to the behavior of the surrounding vehicles, so that the behavior of the surrounding vehicles on the curve is calculated. Even if changes, the travel plan corresponding to it can be made.

It is a block diagram of the automatic driving | operation control apparatus which concerns on this Embodiment. It is a flowchart which shows the flow of the process (learning) in ECU which concerns on 1st Embodiment. It is a flowchart which shows the flow of the process (own vehicle speed pattern generation, preceding vehicle speed pattern generation) in ECU which concerns on 1st Embodiment. It is a flowchart which shows the flow of the process (response at the time of interference, self-vehicle speed pattern regeneration, acceleration / deceleration control) in ECU which concerns on 1st Embodiment. It is a flowchart which shows the flow of the process (learning) in ECU which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the process (own vehicle speed pattern generation | occurrence | production, preceding vehicle speed pattern generation | occurrence | production) in ECU which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the process (The response | compatibility at the time of interference, vehicle speed pattern regeneration, acceleration / deceleration control) in ECU which concerns on 2nd Embodiment. It is a flowchart which shows the flow of the process (the first half of learning) in ECU which concerns on 3rd Embodiment. It is a flowchart which shows the flow of the process (learning second half) in ECU which concerns on 3rd Embodiment. It is a flowchart which shows the flow of the process (own vehicle speed pattern generation | occurrence | production, preceding vehicle speed pattern generation | occurrence | production) in ECU which concerns on 3rd Embodiment. It is a flowchart which shows the flow of the process (The response | compatibility at the time of interference, the own vehicle speed pattern regeneration, acceleration / deceleration control) in ECU which concerns on 3rd Embodiment.

  Hereinafter, an embodiment of a vehicle control device according to the present invention will be described with reference to the drawings. In addition, the same code | symbol is attached | subjected about the element which is the same or it corresponds in each figure, and the overlapping description is abbreviate | omitted.

  In the present embodiment, the vehicle control device according to the present invention is applied to an automatic driving control device mounted on a hybrid vehicle that performs automatic driving. The automatic driving control device according to the present embodiment generates a speed pattern as a travel plan for the host vehicle and controls acceleration / deceleration so as to travel according to the speed pattern in order to control at least the speed of the host vehicle. In the present embodiment, only acceleration / deceleration control in automatic operation control will be described. In the present embodiment, only the speed will be described as the travel plan. However, there are many other travel plans that are necessary for traveling the vehicle such as position (x coordinate, y coordinate), acceleration, yaw angle, and yaw rate. There are parameters.

  In this embodiment, there are three forms, and the first embodiment predicts the speed pattern at the time of curve driving of the preceding vehicle by learning the maximum lateral acceleration and deceleration at the time of curve driving of the preceding vehicle. In the second embodiment, in addition to the maximum lateral acceleration and deceleration at the time of curve driving of the preceding vehicle, the lateral jerk is also learned to predict the speed pattern at the time of curve driving of the preceding vehicle. The embodiment learns the maximum lateral acceleration and deceleration during the curve driving of the preceding vehicle, and learns the correction value for the maximum lateral acceleration and deceleration according to the road shape to obtain the speed pattern during the curve driving of the preceding vehicle. It is a form to predict. Note that when traveling on a curve, it also includes traveling during a deceleration zone before entering the curve.

  In the automatic driving control apparatus according to the present embodiment, a speed pattern is generated in the case where traveling conditions that emphasize fuel efficiency are set. This speed pattern with emphasis on fuel consumption (low fuel consumption) is a speed pattern combining high-efficiency acceleration as acceleration and free run as deceleration. In free run, the engine and the motor are stopped and the vehicle decelerates only with the running resistance.

  With reference to FIG. 1, an automatic driving control apparatus 1 according to the first embodiment will be described. FIG. 1 is a configuration diagram of an automatic driving control apparatus according to the present embodiment.

  The automatic driving control device 1 generates (regenerates) a speed pattern of the host vehicle for the purpose of reducing fuel consumption. In particular, the automatic driving control device 1 learns the maximum lateral acceleration and deceleration at the time of curve traveling of the preceding vehicle so that the preceding vehicle can travel according to the speed pattern even when the preceding vehicle exists ahead of the host vehicle. A speed pattern at the time of curve traveling is predicted, and a speed pattern (particularly at the time of curve traveling) is regenerated according to the predicted speed pattern at the time of curve traveling of the preceding vehicle.

  The automatic driving control device 1 includes a wheel speed sensor 10, a millimeter wave radar 11, a brake actuator 20, a hybrid system 21 (engine 22, motor 23, battery 24), and an ECU [Electronic Control Unit] 31 (preceding vehicle learning unit 31a, A vehicle speed pattern generation unit 31b, a preceding vehicle speed pattern prediction unit 31c, an interference time response unit 31d, a host vehicle speed pattern regeneration unit 31e, and an acceleration / deceleration control unit 31f), and uses information from the navigation system 12. . In the first embodiment, the preceding vehicle learning unit 31a corresponds to the learning unit described in the claims, and the preceding vehicle speed pattern prediction unit 31c serves as the surrounding vehicle behavior prediction unit described in the claims. Correspondingly, the host vehicle speed pattern regeneration unit 31e corresponds to the travel plan generation means described in the claims.

  The wheel speed sensor 10 is a sensor that is provided on each of the four wheels of the vehicle and detects the rotational speed of the wheel (the number of pulses corresponding to the rotation of the wheel). The wheel speed sensor 10 detects the number of rotation pulses of the wheel every predetermined time, and transmits the detected number of wheel rotation pulses to the ECU 31 as a wheel speed signal. The ECU 31 calculates the wheel speed from the rotational speed of each wheel, and calculates the vehicle body speed (vehicle speed) from the wheel speed of each wheel.

  The millimeter wave radar 11 is a radar for detecting a preceding vehicle using millimeter waves. The millimeter wave radar 11 is attached to the front center of the host vehicle. The millimeter wave radar 11 transmits the millimeter wave forward from the own vehicle while scanning the millimeter wave in the left-right direction, and receives the reflected millimeter wave. Then, the millimeter wave radar 11 transmits the millimeter wave transmission / reception information (transmission angle, transmission time, reception angle, reception time, reception intensity, etc. centered on the traveling direction of the host vehicle) to the ECU 31 as a radar signal.

  The navigation system 12 is a system that detects the current position of the host vehicle and provides route guidance to a destination. In particular, the navigation system 12 reads road shape information (including a curve radius and the like) around the current position from the map database, and transmits the road shape information and the detected current position to the ECU 31 as a navigation signal. In the case of a vehicle that does not include a navigation system, a configuration may be provided that includes a map database that stores road shape information, or road shape information may be acquired from infrastructure using road-to-vehicle communication or the like. Further, in the case of a vehicle that does not include a navigation system, a configuration that includes a GPS receiving device in order to acquire the current position may be employed.

  The brake actuator 20 is an actuator that adjusts the brake hydraulic pressure of the wheel cylinder of each wheel. When the brake actuator 20 receives a brake control signal from the ECU 31, the brake actuator 20 operates according to the brake control signal and adjusts the brake hydraulic pressure of the wheel cylinder.

  The hybrid system 21 is a system that controls acceleration / deceleration of the vehicle by combining two drive sources of the engine 22 and the motor 23 singly or in combination. When the hybrid system 21 receives a drive control signal from the ECU 31, the engine 22 and the motor 23 are driven and controlled according to the drive control signal, only the engine 22 is driven, only the motor 23 is driven, and the motor 23 is regeneratively controlled. The engine 22 and the motor 23 are controlled to stop (free-run control). The drive of the engine 22 is controlled by a throttle actuator (not shown) such as an electronic throttle. The drive / regeneration of the motor 23 is controlled by an inverter (not shown). Electric energy is supplied from the battery 24 via the inverter during driving, and the battery 24 is charged with the generated electric energy via the inverter during regeneration. By the way, when performing full regeneration, the full regeneration amount (and hence deceleration) is determined by the maximum power that can be charged to the battery 24 per unit time, and, for example, the deceleration is about -0.2 G due to full regeneration.

  The ECU 31 includes a CPU [Central Processing Unit], various memories, and the like, and performs overall control of the automatic operation control device 1. In the ECU 31, each application program stored in the memory is loaded and executed by the CPU, whereby the preceding vehicle learning unit 31a, the host vehicle speed pattern generation unit 31b, the preceding vehicle speed pattern prediction unit 31c, the interference response unit 31d, A host vehicle speed pattern regeneration unit 31e and an acceleration / deceleration control unit 31f are configured. The ECU 31 receives signals from the sensors 10 and 11 and the navigation system 12 at regular intervals. Then, the ECU 31 performs processing in each unit 31a to 31f based on each received signal, and transmits each control signal to the brake actuator 20 and the hybrid system 21.

  The preceding vehicle learning unit 31a will be described. The preceding vehicle learning unit 31a determines whether or not there is a preceding vehicle, and if there is a preceding vehicle, learns the maximum lateral acceleration and deceleration when the preceding vehicle is traveling on a curve. In this learning, since learning is performed on the same preceding vehicle (and thus the same driver), each learned value is initialized when the preceding vehicle changes.

  Specifically, the preceding vehicle learning unit 31a determines the presence or absence of a preceding vehicle ahead of the host vehicle based on millimeter wave transmission / reception information (particularly, information on a reflection point with received information). Here, the reflection points of the millimeter wave are grouped, and the preceding vehicle can be recognized when the reflection points can be grouped in front of the traveling lane of the host vehicle. The preceding vehicle learning unit 31a uses the information on the reflection point of the group that can be recognized as the preceding vehicle, and based on the speed of the millimeter wave and the time from transmission to reception of the millimeter wave, Calculate the distance. Further, the preceding vehicle learning unit 31a calculates the relative speed between the host vehicle and the preceding vehicle using the frequency change (Doppler effect) of the reflected millimeter wave. The preceding vehicle learning unit 31a calculates the relative direction between the host vehicle and the preceding vehicle based on the millimeter wave reception angle.

  When the preceding vehicle exists, the preceding vehicle learning unit 31a calculates the current position of the preceding vehicle based on the current position of the host vehicle and the relative distance from the preceding vehicle. Further, the preceding vehicle learning unit 31a acquires a curve radius at the current position of the preceding vehicle based on the navigation information.

  The preceding vehicle learning unit 31a determines whether the preceding vehicle enters the curve based on the current position of the preceding vehicle and the curve radius. In this curve entry determination, in order to include the deceleration zone before the curve of the preceding vehicle, the curve (for example, the curve radius is 1000 m or less) is present ahead of the preceding vehicle by a predetermined distance (for example, 300 m ahead). Judged to enter the curve. The preceding vehicle learning unit 31a calculates the deceleration of the preceding vehicle while the preceding vehicle enters the curve. For the deceleration of the preceding vehicle, the speed of the preceding vehicle is calculated based on the speed of the host vehicle and the relative speed with respect to the preceding vehicle at regular time intervals, and the deceleration is calculated from the temporal change in the speed of the preceding vehicle.

The preceding vehicle learning unit 31a determines whether the preceding vehicle is traveling on a straight road (for example, the curve radius is 1000 m or more) based on the curve radius at the current position of the preceding vehicle. When the preceding vehicle is not traveling on the straight road (that is, when the vehicle is traveling between the straight road and the straight road), the preceding vehicle learning unit 31a is based on the speed of the host vehicle and the relative speed with the preceding vehicle. Calculate the speed of the preceding vehicle. Further, the preceding vehicle learning unit 31a calculates the lateral acceleration of the preceding vehicle based on the speed and the curve radius of the preceding vehicle. Assuming that the lateral acceleration Ay, the speed V, and the curve radius R, Ay = V ² / R.

  The preceding vehicle learning unit 31a extracts the maximum lateral acceleration from the calculated lateral acceleration of the preceding vehicle in the current curve, and uses the extracted lateral acceleration as the maximum lateral acceleration of the preceding vehicle in the current curve. . Further, the preceding vehicle learning unit 31a calculates an average value of the deceleration in a deceleration section within a certain value (for example, −0.05G or less) among the deceleration of the preceding vehicle calculated in the current curve. Then, the calculated average value is set as the deceleration of the preceding vehicle on the current curve. The deceleration of the preceding vehicle may be the maximum deceleration among the decelerations of the preceding vehicle calculated in the current curve.

  The preceding vehicle learning unit 31a calculates the average value by adding up the maximum deceleration of each curve that the current preceding vehicle has traveled so far, and sets the calculated average value as the average maximum acceleration of the preceding vehicle. In addition, the preceding vehicle learning unit 31a calculates the average value by summing the decelerations of the curves that the current preceding vehicle has traveled so far, and sets the calculated average value as the average deceleration of the preceding vehicle. The average maximum acceleration and the average deceleration are parameters indicating the running characteristics of the preceding vehicle on the curve, and are parameters for predicting the speed of the preceding vehicle on the curve with high accuracy.

  The own vehicle speed pattern generation unit 31b will be described. The host vehicle speed pattern generation unit 31b generates a speed pattern that serves as a reference for the host vehicle without considering the preceding vehicle. Incidentally, the reference speed pattern is used as it is when traveling on a straight road or on a curved road where no preceding vehicle exists.

  Specifically, the host vehicle speed pattern generation unit 31b uses the road shape information in the navigation information to generate a fuel efficiency-oriented speed pattern consisting of high-efficiency acceleration and free run on condition of upper limit speed and maximum lateral acceleration. To do. The upper limit speed is, for example, a legal speed. As the maximum lateral acceleration, for example, the maximum lateral acceleration desired by the driver of the own vehicle or the maximum lateral acceleration obtained by learning by applying the learning method for the preceding vehicle as described above to the own vehicle. The speed pattern is a speed pattern at a constant time or a speed pattern at a constant minute interval.

An example of a speed pattern generation method will be described. In a region divided into fixed minute intervals in the planned travel route, the maximum speed of the vehicle is calculated using the curve radius and the friction coefficient of the road surface in that region. Assuming that the curve radius is R, the road surface friction coefficient is μ, and the maximum speed is V, V ² = μ × g × R. g is a gravitational acceleration. Then, a steady circle maximum speed pattern including the maximum speed of each region is generated. When a steady circle maximum speed pattern is generated, acceleration is calculated for each adjacent region from the travel start point to the travel end point, and when the calculated acceleration exceeds the upper limit acceleration, the travel end point side where the speed is high The speed is corrected to the low speed side so that it is below the upper limit acceleration. The upper limit acceleration is set to a lower value of a value obtained from the running performance of the vehicle and a value protruding from the friction circle. In addition, a deceleration is calculated for each adjacent area from the travel end point to the travel start point, and if the calculated deceleration exceeds the upper limit deceleration, the upper speed limit on the higher travel start point side is set. Correct to the low speed side so that it is less than the deceleration. The upper limit deceleration is set to a lower value between a value obtained from the running performance of the vehicle and a value protruding from the friction circle. And the speed pattern which consists of the speed of each area | region which satisfy | fills the acceleration / deceleration restriction | limiting condition is produced | generated.

  Another example of the speed pattern generation method is an optimization method. As an optimization method, for example, “Takehiko Fujioka, Daisho Emori: Proceedings of the Society of Automotive Engineers of Japan, Vol. 24, No. 3, July 1993, p106-111“ Theoretical study on the shortest time cornering method ”is disclosed. SCGRA [Sequential Conjugate Gradient Restoration Algorithm] is used. In SCGRA, the convergence calculation is performed based on the steepest descent method until the constraint condition is satisfied, and the convergence calculation is performed based on the conjugate gradient method until the evaluation value of the evaluation function is minimized. The restraint condition is a condition that must be strictly observed when the vehicle is traveling. The evaluation function is a function for evaluating conditions (fuel consumption, etc.) that are important in traveling of the vehicle.

  The own vehicle speed pattern generation unit 31b determines whether or not the average speed of the generated speed pattern of the own vehicle is higher than the average speed of the preceding vehicle. The average speed of the preceding vehicle is a speed obtained by averaging the speed of the preceding vehicle calculated for every predetermined time in a predetermined section. When the average speed of the speed pattern of the host vehicle is higher than the average speed of the preceding vehicle, the host vehicle speed pattern generation unit 31b reduces the upper limit speed condition (for example, − 2 km / h) to regenerate the velocity pattern.

  The preceding vehicle speed pattern prediction unit 31c will be described. When the preceding vehicle enters the curve, the preceding vehicle speed pattern prediction unit 31c predicts and generates a speed pattern at the time of the preceding vehicle traveling on the curve on the condition of the average maximum lateral acceleration and average deceleration of the preceding vehicle acquired by learning. .

  Specifically, the preceding vehicle speed pattern prediction unit 31c determines whether or not the preceding vehicle has already traveled a curve. If the preceding vehicle has never traveled a curve, the average maximum lateral acceleration and the average deceleration for the preceding vehicle are not obtained by learning. Therefore, the preceding vehicle speed pattern prediction unit 31c determines the average maximum lateral acceleration and the average decrease. Set a standard value for each speed. The standard value of the average maximum lateral acceleration is, for example, 0.3G. The standard value of average deceleration is, for example, -0.2G. Each standard value is preset by an actual vehicle experiment or the like.

  The preceding vehicle speed pattern prediction unit 31c determines whether or not the preceding vehicle has entered the curve in the same manner as the preceding vehicle learning unit 31a. When the preceding vehicle enters the curve, the preceding vehicle speed pattern prediction unit 31c uses the road shape information of the navigation information to generate a speed pattern when the preceding vehicle runs on the curve on condition of the average maximum lateral acceleration and the average deceleration. To do. As a speed pattern generation method, the same method described in the host vehicle speed pattern generation unit 31b is used. For example, in the case of the example of the generation method described first, the traveling start point of the speed pattern is the point where the preceding vehicle starts entering the curve, the traveling end point is the end point of the curve, and the initial value of the speed is the preceding vehicle. It is the speed at the point where you entered the curve, the maximum lateral acceleration is the average maximum lateral acceleration, the upper limit deceleration is the average deceleration, and the upper limit acceleration is calculated in the same way as the average deceleration or is important for interference judgment Therefore, a standard value (for example, 0.15 G) is used, and a speed pattern is generated using these parameters.

  The interference response unit 31d will be described. When the preceding vehicle enters the curve, the response unit 31d at the time of interference causes the preceding vehicle to regenerate the speed pattern of the own vehicle according to the speed pattern of the preceding vehicle (because it requires calculation time for regeneration). When it is predicted that the speed pattern of the own vehicle interferes with the speed pattern (a sufficient inter-vehicle time cannot be ensured, there is a risk of a collision, etc.), vehicle control for avoiding the interference is performed. Here, the situation deteriorates in relation to the preceding vehicle while the speed pattern of the own vehicle is being regenerated (for example, the risk of collision increases, the operation of the hydraulic brake (deterioration of fuel consumption with 100% loss)) Vehicle control is performed to minimize the occurrence of

  Specifically, at the time of interference pattern response unit 31d, at each time (or for each minute region) of the speed pattern, based on the speed of the speed pattern of the own vehicle and the speed of the speed pattern of the preceding vehicle, The inter-vehicle time between the vehicle and the preceding vehicle is calculated. Then, the interference time response unit 31d extracts the minimum inter-vehicle time from the calculated inter-vehicle time at each time.

  The interference response unit 31d determines whether the speed pattern of the preceding vehicle and the speed pattern of the host vehicle interfere with each other based on the minimum inter-vehicle time. In this interference determination, the determination is made based on whether or not the minimum inter-vehicle time is equal to or less than the inter-vehicle time set value (for example, 1 second). The inter-vehicle time set value is set in advance by an actual vehicle experiment or the like.

  When the speed pattern of the preceding vehicle and the speed pattern of the host vehicle interfere with each other, the interference response unit 31d determines whether or not the host vehicle is in an acceleration state. When the host vehicle is in an accelerating state, the inter-interaction handling unit 31d transmits a drive control signal indicating free run to the hybrid system 21 in order to perform provisional free run without acceleration. Furthermore, the inter-interaction handling unit 31d determines whether or not the minimum inter-vehicle time is sufficiently small (for example, the minimum inter-vehicle time is less than 0) and there is a risk of collision between the host vehicle and the preceding vehicle. When there is a risk of collision, the interference response unit 31d transmits a drive control signal indicating full regeneration to the hybrid system 21 in order to perform provisional full regeneration (regenerative brake operation) without using the hydraulic brake. .

  The own vehicle speed pattern regeneration unit 31e will be described. When the preceding vehicle enters the curve, the own vehicle speed pattern regeneration unit 31e regenerates the own vehicle speed pattern according to the predicted speed pattern of the preceding vehicle. Here, since speed reduction is a full run (or fuel cut) in a speed pattern that emphasizes fuel consumption, a method of avoiding that the speed pattern of the host vehicle interferes with the speed pattern of the preceding vehicle in order to more efficiently perform the deceleration. As described above, the start point of the deceleration zone is advanced (that is, the point at which the free run is started is set before the point in the speed pattern generated by the host vehicle speed pattern generation unit 31b).

  Specifically, in the own vehicle speed pattern regeneration unit 31e, whether or not the speed pattern of the preceding vehicle and the speed pattern of the own vehicle interfere with each other based on the minimum inter-vehicle time by the same method as in the interference response unit 31d. Determine whether. Until there is no interference between the speed pattern of the preceding vehicle and the speed pattern of the host vehicle, the host vehicle speed pattern regeneration unit 31e gradually decreases the maximum lateral acceleration of the conditions for generating the speed pattern (for example, 0. A speed pattern is generated on the condition of the lowered maximum acceleration. By reducing the maximum lateral acceleration in this way, the deceleration start point of the host vehicle can be advanced in the speed pattern, and the preceding vehicle can be decelerated in advance corresponding to the deceleration before entering the curve.

  The acceleration / deceleration control unit 31f will be described. The acceleration / deceleration control unit 31f performs acceleration / deceleration control according to the generated speed pattern of the host vehicle. Specifically, the acceleration / deceleration control unit 31f drives the drive control signal (target acceleration) based on the deviation between the current speed of the host vehicle and the speed at the corresponding time of the speed pattern so as to travel according to the speed pattern of the host vehicle. , Free run, target regeneration amount, etc.) or brake control signal (target deceleration), and each signal is transmitted to the brake actuator 20 and the hybrid system 21, respectively.

  With reference to FIG. 1, the operation | movement in the automatic driving | operation control apparatus 1 is demonstrated. In particular, the processing in the ECU 31 will be described with reference to FIGS. FIG. 2 is a flowchart showing a flow of processing (learning) in the ECU 31. FIG. 3 is a flowchart showing a flow of processing (vehicle speed pattern generation, preceding vehicle speed pattern generation) in the ECU 31. FIG. 4 is a flowchart showing a flow of processing (corresponding to interference, regeneration of own vehicle speed pattern, acceleration / deceleration control) in the ECU 31.

  The wheel speed sensor 10 detects the number of rotation pulses of the wheel at regular time intervals and transmits a wheel speed signal to the ECU 31. The ECU 31 receives the wheel speed signal and calculates the speed of the host vehicle from the rotational speed of each wheel. The millimeter wave radar 11 transmits the millimeter wave forward while scanning the millimeter wave in the left-right direction at regular intervals, receives the reflected millimeter wave, and transmits the radar signal to the ECU 31. The ECU 31 receives radar signals and acquires millimeter wave transmission / reception information (transmission angle, transmission time, reception angle, reception time, reception intensity, etc.). The navigation system 12 detects the current position at regular intervals, reads road shape information around the current position from the map database, and transmits the road shape information and current position information to the ECU 31 as navigation signals. The ECU 31 receives the navigation signal and acquires road shape information and current position information.

  The ECU 31 repeats the following processing at regular intervals. First, the ECU 31 determines whether there is a preceding vehicle ahead of the host vehicle based on the millimeter wave transmission / reception information (S10). When it is determined in S10 that there is no preceding vehicle, the ECU 31 skips to the process of S22.

  When it is determined in S10 that the preceding vehicle exists, the ECU 31 calculates the relative distance between the own vehicle and the preceding vehicle based on the millimeter wave transmission / reception information. Then, the ECU 31 calculates the current position of the preceding vehicle based on the current position of the host vehicle obtained from the navigation information and the relative distance from the preceding vehicle (S11). Further, the ECU 31 acquires the curve radius at the current position of the preceding vehicle from the navigation information (S12).

  The ECU 31 determines whether or not the preceding vehicle has entered the curve based on the current position of the preceding vehicle and the curve radius (S13). If it is determined in S13 that the preceding vehicle has entered the curve, the ECU 31 calculates the deceleration of the preceding vehicle (S14).

  The ECU 31 determines whether or not the preceding vehicle is going straight on the basis of the curve radius of the preceding vehicle (S15). If it is determined in S15 that the preceding vehicle is traveling straight, the ECU 31 skips to the process of S22.

  If it is determined in S15 that the preceding vehicle is not going straight (that is, a curve), the ECU 31 calculates the speed of the preceding vehicle based on the speed of the host vehicle and the relative speed with the preceding vehicle (S16). Further, the ECU 31 calculates the lateral acceleration of the preceding vehicle based on the speed of the preceding vehicle and the curve radius (S17).

  Then, the ECU 31 extracts the maximum lateral acceleration of the preceding vehicle from the maximum lateral acceleration at each time in the current curve (S18). Further, the ECU 31 calculates a deceleration obtained by averaging the deceleration at each time in the current curve (S19).

  Further, the ECU 31 calculates an average maximum lateral acceleration obtained by averaging the maximum lateral accelerations of the curves that the preceding vehicle has traveled so far (S20). Further, the ECU 31 calculates an average deceleration obtained by averaging the deceleration of each curve that the preceding vehicle has traveled so far (S21).

  The ECU 31 determines whether or not the current preceding vehicle has traveled the curve even once (S22). If it is determined in S22 that the current preceding vehicle has never traveled the curve, the ECU 31 sets standard values as the average maximum lateral acceleration and average deceleration of the preceding vehicle, respectively (S23).

  The ECU 31 uses the road shape information to generate a speed pattern emphasizing fuel efficiency consisting of high-efficiency acceleration and free run on condition of an upper limit speed and maximum lateral acceleration (S24). Then, the ECU 31 determines whether or not the average speed of the speed pattern of the host vehicle is higher than the average speed of the preceding vehicle (S25). If it is determined in S25 that the average speed of the speed pattern of the host vehicle is higher than the average speed of the preceding vehicle, the ECU 31 lowers the upper limit speed of the speed pattern generation condition (S26) and generates the speed pattern again. (S24).

  If it is determined in S25 that the average speed of the speed pattern of the host vehicle is not higher than the average speed of the preceding vehicle, the ECU 31 determines whether the preceding vehicle enters the curve based on the current position of the preceding vehicle and the curve radius. Is determined (S27). If it is determined in S27 that the preceding vehicle has not entered the curve, the ECU 31 skips to the process of S38.

  If it is determined in S27 that the preceding vehicle has entered the curve, the ECU 31 uses the road shape information to determine the speed pattern when the preceding vehicle is traveling on the curve on the condition of the average maximum lateral acceleration and the average deceleration of the preceding vehicle. Prediction is made (S28).

  The ECU 31 calculates the inter-vehicle time between the host vehicle and the preceding vehicle based on the respective speeds of the speed pattern of the host vehicle and the speed pattern of the preceding vehicle at each time, and extracts the minimum inter-vehicle time from the calculated time. (S29). Then, the ECU 31 determines whether or not the speed pattern of the host vehicle and the speed pattern of the preceding vehicle interfere with each other based on the minimum inter-vehicle time (S30). When it determines with not interfering in S30, ECU31 skips to the process of S38.

  When it determines with interfering in S30, ECU31 determines whether the own vehicle is an acceleration state (S31). When it is determined in S31 that the host vehicle is in the acceleration state, the ECU 31 transmits a drive control signal indicating the free run to the hybrid system 21 in order to perform a temporary free run (S32). When this drive control signal is received, the hybrid system 21 stops the engine 22 and the motor 23. Further, the ECU 31 determines whether or not there is a risk of collision between the host vehicle and the preceding vehicle (S33). If it is determined in S33 that there is a risk of collision, the ECU 31 transmits a drive control signal indicating full regeneration to the hybrid system 21 in order to perform provisional full regeneration (S34). When this drive control signal is received, the hybrid system 21 performs full regeneration of the motor 23 according to the charging capacity of the battery 24 (regenerative braking operation), and charges the battery 24 with the generated electric power.

  In order to avoid interference between the speed pattern of the host vehicle and the speed pattern of the preceding vehicle, the ECU 31 reduces the maximum lateral acceleration of the speed pattern generation condition (S35), and sets the speed pattern of the host vehicle on the condition of the maximum lateral acceleration. It is regenerated (S36). Then, the ECU 32 calculates the minimum inter-vehicle time by the method similar to the above using the regenerated speed pattern of the own vehicle, and the speed pattern of the own vehicle and the speed pattern of the preceding vehicle are calculated based on the minimum inter-vehicle time. It is determined whether or not interference occurs (S37). If it is determined that the interference occurs in S37, the ECU 31 further lowers the maximum lateral acceleration of the speed pattern generation condition (S35), and regenerates the host vehicle on the condition of the maximum lateral acceleration (S36).

  If it is determined that there is no interference in S37, the ECU 31 drives the brake control signal or drive based on the deviation between the current speed of the host vehicle and the speed at the corresponding time of the speed pattern so as to travel according to the speed pattern of the host vehicle. A control signal is generated, and each control signal is transmitted to the brake actuator 20 or the hybrid system 21. When receiving the brake control signal, the brake actuator 20 operates according to the brake control signal and adjusts the brake hydraulic pressure of the wheel cylinder. When the hybrid system 21 receives a drive control signal, the hybrid system 21 controls the drive of the engine 22 and / or the drive / regeneration of the motor based on the drive control signal.

  According to this automatic driving control device 1, the speed pattern of the preceding vehicle at the time of curve traveling is predicted based on the maximum lateral acceleration and deceleration, and the speed pattern of the own vehicle is regenerated according to the speed pattern of the preceding vehicle. Thus, even when the speed of the preceding vehicle changes due to the curve, an optimum speed pattern of the host vehicle corresponding to the speed can be established. For example, even when the preceding vehicle decelerates toward the curve approach, it is possible to reduce the speed before the preceding vehicle decelerates by performing acceleration / deceleration control according to the regenerated vehicle speed pattern. There is no need to perform unnecessary deceleration that is not in the speed pattern, and the vehicle can travel according to the speed pattern. As a result, it is possible to achieve conditions that emphasize fuel efficiency when setting the speed pattern.

  In addition, according to the automatic driving control device 1, by learning the maximum lateral acceleration and deceleration at the time of curve traveling for the preceding vehicle, it is possible to obtain the curve traveling characteristics of each preceding vehicle (individual driver). In addition, it is possible to predict the speed pattern of the preceding vehicle during curve traveling with high accuracy.

  Further, according to the automatic driving control device 1, when the speed pattern of the host vehicle and the preceding vehicle interferes until the speed pattern of the host vehicle is regenerated, the corresponding vehicle control is performed. Interference with the vehicle can be avoided and safety can be ensured. Furthermore, since a large deceleration that requires a hydraulic brake is prevented by free running or regenerative frames, deterioration of fuel consumption can be minimized.

  Further, according to the automatic driving control device 1, when the speed pattern of the host vehicle and the preceding vehicle interferes, the maximum lateral acceleration is decreased and the speed pattern of the host vehicle is regenerated, thereby reducing the speed of the entire speed pattern. Without this, it is possible to cope with deceleration on the curve of the preceding vehicle by advancing the deceleration start point of the host vehicle.

  With reference to FIG. 1, an automatic operation control device 2 according to a second embodiment will be described. Compared with the automatic driving control apparatus 1 according to the first embodiment, the automatic driving control apparatus 2 obtains a differential value of the maximum lateral acceleration (lateral jerk) in order to obtain a more accurate speed pattern in the curve of the preceding vehicle. ) Is also learned, and the speed pattern at the time of curve driving of the preceding vehicle is predicted in consideration of the lateral jerk. The automatic operation control device 2 is different from the configuration of the automatic operation control device 1 only in the processing in the ECU 32, and only the ECU 32 will be described.

  Regarding horizontal jerk, it is known that variation is small in many situations for the same individual. This is probably because the lateral acceleration (= traveling speed) depends on the road conditions, whereas the lateral jerk depends on the driver's turning speed.

  The ECU 32 includes a CPU, various memories, and the like, and comprehensively controls the automatic operation control device 2. In the ECU 32, each application program stored in the memory is loaded and executed by the CPU, whereby the preceding vehicle learning unit 32a, the host vehicle speed pattern generation unit 32b, the preceding vehicle speed pattern prediction unit 32c, the interference response unit 32d, A host vehicle speed pattern regeneration unit 32e and an acceleration / deceleration control unit 32f are configured. In the second embodiment, the preceding vehicle learning unit 32a corresponds to the learning unit described in the claims, and the preceding vehicle speed pattern prediction unit 32c serves as the surrounding vehicle behavior prediction unit described in the claims. The own vehicle speed pattern regeneration unit 32e corresponds to the travel plan generation unit described in the claims.

  In addition, about the own vehicle speed pattern generation part 32b, the interference time response part 32d, the own vehicle speed pattern regeneration part 32e, and the acceleration / deceleration control part 32f, the own vehicle speed pattern generation part 31b according to the first embodiment, the interference The same processing as that performed by the hour handling unit 31d, the host vehicle speed pattern regeneration unit 31e, and the acceleration / deceleration control unit 31f is performed, and thus the description thereof is omitted.

  The preceding vehicle learning unit 32a will be described. The preceding vehicle learning unit 32a performs the same processing as the preceding vehicle learning unit 31a according to the first embodiment, and also performs the following processing. The preceding vehicle learning unit 32a learns a lateral jerk in addition to the maximum lateral acceleration and deceleration when the preceding vehicle is traveling on a curve. Here, only the side jerk learning process will be described.

Specifically, when the preceding vehicle learning unit 32a calculates the lateral acceleration Ay (n) at the current time, the difference between the lateral acceleration Ay (n) at the current time and the lateral acceleration Ay (n-1) at the previous time. Is calculated, and the value is defined as a horizontal jerk Jy (n). Then, the preceding vehicle learning unit 32a calculates the average value of the lateral jerk in the section of the lateral jerk that is equal to or greater than a certain value (for example, 0.1 m / s ³ or more) among the lateral jerk of the preceding vehicle calculated in the current curve. The calculated average value is set as the lateral jerk of the preceding vehicle on the current curve. Further, the preceding vehicle learning unit 32a calculates the average value by summing the lateral jerks of the curves that the current preceding vehicle has traveled so far, and sets the calculated average value as the average lateral jerk Jave of the preceding vehicle. .

  The preceding vehicle speed pattern prediction unit 32c will be described. When the preceding vehicle enters the curve, the preceding vehicle speed pattern predicting unit 32c predicts and generates a speed pattern at the time of the preceding vehicle traveling on the curve on the condition that the average maximum lateral acceleration and the average deceleration of the preceding vehicle acquired by learning are used as conditions. The predicted speed pattern is corrected based on the average lateral jerk Jave of the preceding vehicle acquired by learning.

Specifically, the preceding vehicle speed pattern prediction unit 32c determines whether or not the preceding vehicle has already traveled a curve. If the preceding vehicle has never traveled a curve, the average maximum lateral acceleration, average deceleration, and average lateral jerk for the preceding vehicle are not obtained by learning, so the preceding vehicle speed pattern predicting unit 32c determines the average maximum lateral Standard values are set for acceleration, average deceleration, and average lateral jerk, respectively. The standard value of the average lateral jerk is, for example, 1 m / s ³ . Then, the preceding vehicle speed pattern predicting unit 32c performs the same processing as the preceding vehicle speed pattern predicting unit 31c according to the first embodiment on the condition that the preceding vehicle is in a curve traveling on the condition of the average maximum lateral acceleration and the average deceleration. Generate a velocity pattern.

  When the speed pattern at the time of curve driving of the preceding vehicle is generated, the preceding vehicle speed pattern prediction unit 32c uses the speed V (n) at each time for all the times of the speed pattern of the preceding vehicle, and the lateral acceleration at each time. Ay (n) is calculated respectively. Then, the preceding vehicle speed pattern predicting unit 32c calculates the absolute value of the difference between the lateral acceleration Ay (n) at each time and the lateral acceleration Ay (n-1) at the previous time, and the speed pattern of the preceding vehicle. The horizontal jerk Jy (n) for all the times is calculated.

  The preceding vehicle speed pattern prediction unit 32c determines whether or not the calculated lateral jerk Jy (n) at all times is smaller than the average lateral jerk Jave obtained by learning. If the horizontal jerk Jy (n) at all times is not smaller than the average horizontal jerk Jave, Ay (n) so that the horizontal jerk Jy (n) at all times falls within the average horizontal jerk Jave obtained by learning. ) And Ay (n−1), the larger value (the larger one of V (n) and V (n−1)) is lowered to correct the speed pattern of the preceding vehicle.

  Specifically, the preceding vehicle speed pattern prediction unit 32c determines whether or not the vehicle is accelerating (V (n-1) = <V (n)) for each time n in the speed pattern of the preceding vehicle. When the vehicle is accelerating at time n, the preceding vehicle speed pattern prediction unit 32c calculates lateral acceleration Ay (n) at the time n = lateral acceleration Ay (n-1) at the previous time n−1 + average lateral jerk Jave. To do. Further, the preceding vehicle speed pattern prediction unit 32c uses the calculated lateral acceleration Ay (n) at time n to use the speed V (n) at time n = sqrt (lateral acceleration Ay (n) at time n × curve radius. R) is calculated. On the other hand, when the vehicle is decelerating, the preceding vehicle speed pattern prediction unit 32c calculates lateral acceleration Ay (n-1) at time n-1 before time n = lateral acceleration Ay (n) at time n + average lateral jerk Jave. To do. Further, the preceding vehicle speed pattern prediction unit 32c uses the calculated lateral acceleration Ay (n-1) at the previous time n-1 to use the speed V (n-1) = sqrt (the previous time at the previous time n-1). The lateral acceleration Ay (n−1) × curve radius R) of n−1 is calculated. This process is performed for all times n of the speed pattern of the preceding vehicle.

  Further, in the preceding vehicle speed pattern prediction unit 32c, there is a possibility that the acceleration / deceleration restriction condition of the speed pattern may not be satisfied as a result of correcting the speed V (n) at each time by the above processing. The speed V (n) at all times n in the speed pattern of the preceding vehicle is re-corrected so as to satisfy. For example, the re-correction based on the acceleration / deceleration restriction condition is re-corrected by the process of correcting the speed exceeding the upper limit acceleration and the upper limit deceleration with respect to the steady circle maximum speed pattern described as an example of the speed pattern generation method. Do.

  With reference to FIG. 1, the operation | movement in the automatic driving | operation control apparatus 2 is demonstrated. In particular, the processing in the ECU 32 will be described with reference to FIGS. FIG. 5 is a flowchart showing a flow of processing (learning) in the ECU 32. FIG. 6 is a flowchart showing a flow of processing (vehicle speed pattern generation, preceding vehicle speed pattern generation) in the ECU 32. FIG. 7 is a flowchart showing a flow of processing (corresponding to interference, regeneration of own vehicle speed pattern, acceleration / deceleration control) in the ECU 32.

  About the wheel speed sensor 10, the millimeter wave radar 11, and the navigation system 12, the same operation | movement demonstrated with the automatic driving | operation control apparatus 1 which concerns on 1st Embodiment is performed.

  The ECU 32 repeats the following processing at regular intervals. First, in ECU32, about the process of S40-S47, the process similar to S10-S17 in ECU31 which concerns on 1st Embodiment is performed. When the lateral acceleration Ay (n) of the preceding vehicle is calculated, the ECU 32 uses the lateral acceleration Ay (n) at the time n and the lateral acceleration Ay (n-1) at the previous time n−1 to calculate the lateral jerk Jy ( n) is calculated (S48).

  Then, the ECU 32 extracts the maximum lateral acceleration of the preceding vehicle in the current curve by the same processing as S18 and S19 in the ECU 31 according to the first embodiment (S49), and in the current curve A deceleration obtained by averaging the deceleration is calculated (S50). Further, the ECU 32 calculates a lateral jerk obtained by averaging the lateral jerks in the current curve (S51).

  Further, the ECU 32 calculates the average maximum lateral acceleration of the preceding vehicle at each curve so far by the same processing as S20 and S21 in the ECU 31 according to the first embodiment (S52). The average deceleration of the preceding vehicle on the curve is calculated (S53). Further, the ECU 32 calculates an average lateral jerk Jave obtained by averaging the lateral jerk of the preceding vehicle at each curve that the preceding vehicle has traveled so far (S54).

  The ECU 32 determines whether or not the current preceding vehicle has already traveled the curve even once (S55). If it is determined in S55 that the current preceding vehicle has never traveled the curve, the ECU 31 sets standard values for the average maximum lateral acceleration, average deceleration, and average lateral jerk of the preceding vehicle (S56). ).

  In ECU32, about the process of S57-S59, the process similar to S24-S26 in ECU31 which concerns on 1st Embodiment is performed, and the speed pattern of the own vehicle is produced | generated.

  The ECU 32 determines whether the preceding vehicle enters the curve based on the current position of the preceding vehicle and the curve radius (S60). If it is determined in S60 that the preceding vehicle has not entered the curve, the ECU 32 skips to the process of S79.

  If it is determined in S60 that the preceding vehicle has entered the curve, the ECU 32 uses the road shape information to determine the speed pattern when the preceding vehicle is traveling on the curve on the condition of the average maximum lateral acceleration and average deceleration of the preceding vehicle. Prediction is performed (S61).

  The ECU 32 calculates the lateral acceleration Ay (n) at each time of the preceding vehicle using the speed V (n) at each time of the speed pattern of the preceding vehicle (S62). Then, the ECU 32 calculates the lateral jerk Jy (n) at each time of the preceding vehicle using the calculated lateral acceleration Ay (n) at each time and the lateral acceleration Ay (n-1) at the previous time. (S63). When the horizontal jerk Jy (n) at all times is calculated, the ECU 32 determines whether or not the horizontal jerk Jy (n) at all times is smaller than the average horizontal jerk Jave (S64). When it is determined in S64 that the horizontal jerk Jy (n) at all times is smaller than the average horizontal jerk Jave, the ECU 32 skips to the process of S70.

  If it is determined in S64 that the lateral jerk Jy (n) at all times is not smaller than the average lateral jerk Jave, the ECU 32 determines whether or not the time n in the speed pattern of the preceding vehicle is accelerating (S65). . If it is determined in S65 that the time n is accelerating, the ECU 32 reduces the horizontal jerk J (n) at the time n using the average horizontal jerk Jave, and uses the reduced horizontal jerk J (n) to determine the time n. Is reduced (S66). On the other hand, if it is determined in S65 that the time n is not accelerating, the ECU 32 uses the average lateral jerk Jave to reduce the lateral jerk J (n-1) at the previous time n-1 and reduces the lateral jerk J that has been reduced. The speed V (n-1) at the previous time n-1 is reduced using (n-1) (S67). Then, the ECU 32 determines whether or not processing for all times n has been completed (S68). If it is determined in S68 that the process for all the times has not been completed, the ECU 32 returns to the process of S65 and performs the process for the next time. On the other hand, when it is determined in S68 that the processing for all the times has been completed, the ECU 32 re-corrects the corrected speeds V (n) for all the times so as to satisfy the acceleration restriction condition (S69). Here, the final speed pattern of the preceding vehicle is generated.

  In ECU32, about the process of S70-S79, the process similar to S29-S38 in ECU31 which concerns on 1st Embodiment is performed. When each control signal of the ECU 32 is received, the brake actuator 20 and the hybrid system 21 perform the same operation as described in the automatic operation control device 1 according to the first embodiment.

  The automatic operation control device 2 has the same effects as the automatic operation control device 1 according to the first embodiment, and also has the following effects. According to the automatic driving control device 2, it is possible to obtain more accurate curve traveling characteristics of each preceding vehicle (individual driver) by learning the lateral jerk during the curve traveling of the preceding vehicle. It is possible to predict with higher accuracy than the speed pattern at the time of running the curve. As a result, a more optimal speed pattern of the host vehicle can be set, and effects such as improvement in fuel consumption of the host vehicle can be acquired to the maximum. For example, even when the preceding vehicle decelerates unexpectedly on a curve, the speed can be reduced correspondingly and fuel consumption deterioration can be avoided.

  With reference to FIG. 1, an automatic driving control device 3 according to a third embodiment will be described. Compared to the automatic driving control device 1 according to the first embodiment, the automatic driving control device 3 is configured to obtain a road pattern after a curve (in particular, a curve) in order to obtain a more accurate speed pattern in the curve of the preceding vehicle. It also learns the correction values for lateral acceleration and deceleration for each road width that enters later, and corrects the lateral acceleration and deceleration according to the shape of the road after the curve to predict the speed pattern of the preceding vehicle when driving a curve. Is different. The automatic operation control device 3 is different from the configuration of the automatic operation control device 1 only in the processing in the ECU 33, and only the ECU 33 will be described.

  In addition, it is considered that the lateral acceleration and deceleration on the curve tend to be determined depending on the road shape after the curve. For example, when turning right or left at an intersection and entering a narrow road ahead, the maximum lateral acceleration in the curve tends to be low because it is desired to reduce the speed. When approaching, the maximum lateral acceleration in the curve tends to increase. Assuming this tendency, lateral acceleration and deceleration depending on the road width of the destination after the curve (road width that is difficult to face, road that can be face-to-face without a center line, road with a center line, etc.) Learn the correction value.

  The ECU 33 includes a CPU, various memories, and the like, and performs overall control of the automatic operation control device 3. In the ECU 33, each application program stored in the memory is loaded and executed by the CPU, whereby the preceding vehicle learning unit 33a, the host vehicle speed pattern generation unit 33b, the preceding vehicle speed pattern prediction unit 33c, the interference response unit 33d, A host vehicle speed pattern regeneration unit 33e and an acceleration / deceleration control unit 33f are configured. In the third embodiment, the preceding vehicle learning unit 33a corresponds to the learning unit described in the claims, and the preceding vehicle speed pattern prediction unit 33c serves as the surrounding vehicle behavior prediction unit described in the claims. Correspondingly, the host vehicle speed pattern regeneration unit 33e corresponds to the travel plan generation means described in the claims.

  In addition, about the own vehicle speed pattern generation unit 33b, the interference response unit 33d, the own vehicle speed pattern regeneration unit 33e, and the acceleration / deceleration control unit 33f, the own vehicle speed pattern generation unit 31b according to the first embodiment, the interference The same processing as that performed by the hour handling unit 31d, the host vehicle speed pattern regeneration unit 31e, and the acceleration / deceleration control unit 31f is performed, and thus the description thereof is omitted.

  The preceding vehicle learning unit 33a will be described. The preceding vehicle learning unit 33a performs the same processing as the preceding vehicle learning unit 31a according to the first embodiment, and also performs the following processing. The preceding vehicle learning unit 33a learns the lateral acceleration correction value and the deceleration correction value in the curve of the preceding vehicle. If there is a road type after the curve, the lateral acceleration correction value and the deceleration correction according to the road type. The value is learned, and further, the road width lateral acceleration correction coefficient and the road width deceleration coefficient corresponding to the road width after the curve are learned. Here, only learning of the correction value and the correction coefficient will be described.

  Specifically, the preceding vehicle learning unit 33a divides the maximum lateral acceleration obtained from the current curve by the average maximum lateral acceleration obtained from all the previous curves, and calculates the lateral acceleration correction value in the current curve. calculate. Further, the preceding vehicle learning unit 33a divides the deceleration obtained from the current curve by the average deceleration obtained from all the previous curves, and calculates the deceleration correction value for the current curve. The preceding vehicle learning unit 33a stores the lateral acceleration correction value and the deceleration correction value calculated together with the current position of the current curve obtained from the navigation information in a storage device such as a hard disk. Here, the maximum lateral acceleration or deceleration in each curve is divided by the average value in all the curves, thereby obtaining a correction value indicating the characteristic in each curve.

  The preceding vehicle learning unit 33a determines whether or not a road type after a curve (a road that is difficult to face, a road that can be faced without a center line, a road with a center line, etc.) can be acquired. As a method for acquiring the road type after the curve, the road type is acquired from the map database of the navigation system 12 or acquired by infrastructure information using road-to-vehicle communication. When the road type after the curve can be acquired, the preceding vehicle learning unit 33a calculates the average value by calculating the lateral acceleration correction values for each curve so far for the road type, and the previous curve Aggregate the deceleration correction values and calculate the average value. The preceding vehicle learning unit 33a stores the lateral acceleration correction value (average value) and the deceleration correction value (average value) calculated in association with the road type in the storage device. Here, the correction value for each road type after the curve is obtained.

  The preceding vehicle learning unit 33a acquires the road width of the road after the curve. As a method for acquiring the road width, the road width is acquired from a map database of the navigation system 12 or measured by a front side sensor of the host vehicle. The preceding vehicle learning unit 33a calculates the road width lateral acceleration coefficient by dividing the acquired road width by the lateral acceleration correction value obtained from the current curve, and reduces the road width obtained from the current curve. Divide by the speed correction value to calculate the road width deceleration coefficient. The preceding vehicle learning unit 33a calculates the average value by calculating the road width lateral acceleration coefficient of all the curves that the preceding vehicle has traveled so far, and also calculates the average value of all the curves that the preceding vehicle has traveled so far. The average value is calculated by summing up the road width deceleration coefficients. The preceding vehicle learning unit 33a stores the calculated road width lateral acceleration coefficient (average value) and road width deceleration coefficient (average value) in the storage device. Here, it is assumed that the road width after the curve and the lateral acceleration and deceleration at the curve are in a proportional relationship, and the proportional coefficient is obtained as a correction coefficient.

  The preceding vehicle speed pattern prediction unit 33c will be described. In the preceding vehicle speed pattern prediction unit 33c, when the preceding vehicle enters the curve, learning is performed using any one of the correction value for the current curve, the correction value for the road type after the curve, or the correction coefficient for the road width after the curve. The acquired average maximum lateral acceleration and average deceleration of the preceding vehicle are corrected, and a speed pattern during curve traveling of the preceding vehicle is predicted and generated on the condition of the corrected average maximum lateral acceleration and average deceleration of the preceding vehicle.

  Specifically, when the preceding vehicle speed pattern predicting unit 33c determines that the preceding vehicle enters the curve by the same process as the preceding vehicle speed pattern predicting unit 31c according to the first embodiment, the curve is determined from the navigation information. The lateral acceleration correction value and deceleration correction value for the current curve are stored in the storage device by comparing the acquired current position with the current position of each curve stored in the storage device. It is determined whether or not it has been done. When the lateral acceleration correction value and the deceleration correction value for the current curve are stored, the preceding vehicle speed pattern prediction unit 33c reads the lateral acceleration correction value and the deceleration correction value from the storage device. Then, the preceding vehicle speed pattern predicting unit 33c adds the lateral acceleration correction value to the average maximum lateral acceleration, corrects the average maximum lateral acceleration, adds the deceleration correction value to the average deceleration, and calculates the average deceleration. Correct.

  When the lateral acceleration correction value and the deceleration correction value for the current curve are not stored (that is, when the preceding vehicle travels for the first time), the preceding vehicle speed pattern prediction unit 33c causes the road to enter after the curve. It is determined whether the type can be acquired. When the road type after the curve can be acquired, the preceding vehicle speed pattern prediction unit 33c reads the lateral acceleration correction value and the deceleration correction value associated with the same road type from the storage device. Then, the preceding vehicle speed pattern predicting unit 33c adds the lateral acceleration correction value to the average maximum lateral acceleration, corrects the average maximum lateral acceleration, adds the deceleration correction value to the average deceleration, and calculates the average deceleration. Correct.

  When the road type after the curve cannot be acquired, the preceding vehicle speed pattern prediction unit 33c acquires the road width of the road that enters after the curve. The preceding vehicle speed pattern prediction unit 33c reads the road width lateral acceleration coefficient and the road width deceleration coefficient from the storage device. Then, the preceding vehicle speed pattern predicting unit 33c adds the road width lateral acceleration coefficient to the acquired road width to calculate a lateral acceleration correction value, adds the road width deceleration coefficient to the acquired road width, A speed correction value is calculated. Further, the preceding vehicle speed pattern predicting unit 33c adds the calculated lateral acceleration correction value to the average maximum lateral acceleration, corrects the average maximum lateral acceleration, and adds the calculated deceleration correction value to the average deceleration. , Correct the average deceleration.

  Then, the preceding vehicle speed pattern prediction unit 32c performs the same process as the preceding vehicle speed pattern prediction unit 31c according to the first embodiment on the condition of the corrected average maximum acceleration and average deceleration, as a condition. Generate a speed pattern when driving.

  With reference to FIG. 1, the operation | movement in the automatic driving | operation control apparatus 3 is demonstrated. In particular, the processing in the ECU 33 will be described with reference to FIGS. FIG. 8 is a flowchart showing the flow of processing (the first half of learning) in the ECU 33. FIG. 9 is a flowchart showing a flow of processing (second half of learning) in the ECU 33. FIG. 10 is a flowchart showing a flow of processing in the ECU 33 (vehicle speed pattern generation, preceding vehicle speed pattern generation). FIG. 11 is a flowchart showing the flow of processing (corresponding to interference, regeneration of own vehicle speed pattern, acceleration / deceleration control) in the ECU 33.

  About the wheel speed sensor 10, the millimeter wave radar 11, and the navigation system 12, the same operation | movement demonstrated with the automatic driving | operation control apparatus 1 which concerns on 1st Embodiment is performed.

  The ECU 33 repeats the following processing at regular intervals. First, the ECU 33 performs the same processes as S10 to S21 in the ECU 31 according to the first embodiment for the processes of S80 to S91, and learns the maximum lateral acceleration and deceleration of the preceding vehicle on the curve.

  The ECU 33 divides the maximum lateral acceleration on the current curve by the average maximum lateral acceleration, and calculates the lateral acceleration correction value of the preceding vehicle on the current curve (S92). Further, the ECU 33 divides the deceleration on the current curve by the average deceleration, and calculates the deceleration correction value for the preceding vehicle on the current curve (S93). Then, the ECU 33 stores the lateral acceleration correction value and the deceleration correction value in the storage device in association with the current position of the current curve (S94).

  The ECU 33 determines whether or not the road type after the curve can be acquired (S95). When the road type after the curve can be acquired, the ECU 33 calculates the average value of the lateral acceleration correction value for each curve so far for the road type, and averages the deceleration correction value for each curve so far. A value is calculated (S96). Then, the ECU 33 stores the lateral acceleration correction value (average value) and the deceleration correction value (average value) in the storage device in association with the road type (S96).

  Further, the ECU 33 acquires the road width after the curve. Then, the ECU 33 divides the road width by the lateral acceleration correction value to calculate the road width lateral acceleration coefficient, and also divides the road width by the deceleration correction value to calculate the road width deceleration coefficient (S97). . Further, the ECU 33 calculates the average value of the road width lateral acceleration coefficient for each curve so far, and calculates the average value of the road width deceleration coefficient for each curve so far (S98). Then, the ECU 33 stores the road width lateral acceleration coefficient (average value) and the road width deceleration coefficient (average value) in the storage device (S98).

  The ECU 33 performs the same processes as S22 and S23 in the ECU 31 according to the first embodiment for the processes of S99 and S100, and the average maximum lateral acceleration when the preceding vehicle has not traveled the curve even once. Set a standard value for the average deceleration. Moreover, in ECU33, about the process of S101-S103, the process similar to S24-S26 in ECU31 which concerns on 1st Embodiment is performed, and the speed pattern of the own vehicle is produced | generated.

  The ECU 33 determines whether or not the preceding vehicle has entered the curve based on the current position of the preceding vehicle and the curve radius (S104). If it is determined in S104 that the preceding vehicle has not entered the curve, the ECU 33 skips to S120.

  If it is determined in S104 that the preceding vehicle has entered the curve, the ECU 33 determines whether the storage device has a lateral acceleration correction value and a deceleration correction value for the current curve (S105). When it is determined in S105 that the lateral acceleration correction value and the deceleration correction value are present, the ECU 33 corrects the average maximum lateral acceleration using the lateral acceleration correction value and uses the deceleration correction value to average deceleration. Is corrected (S106).

  If it is determined in S105 that there is no lateral acceleration correction value and deceleration correction value for the current curve, the ECU 33 determines whether or not the road type after the curve can be acquired (S107). If it is determined in S107 that the road type after the curve can be acquired, the ECU 33 reads the lateral acceleration correction value and the deceleration correction value corresponding to the road type from the storage device (S108). Then, the ECU 33 corrects the average maximum lateral acceleration using the read lateral acceleration correction value, and corrects the average deceleration using the read deceleration correction value (S108).

  If it is determined in S107 that the road type after the curve cannot be acquired, the ECU 33 acquires the road width after the curve, and reads the road width lateral acceleration coefficient and the road width deceleration coefficient from the storage device (S109). The ECU 33 corrects the average maximum lateral acceleration using the read road width lateral acceleration coefficient and the road width after the current curve, and uses the read deceleration correction value and the road width after the current curve. The average deceleration is corrected (S109).

  Then, the ECU 33 predicts a speed pattern during the curve traveling of the preceding vehicle on the condition of the corrected average maximum lateral acceleration and average deceleration (S110).

  In ECU33, about the process of S111-S120, the process similar to S29-S38 in ECU31 which concerns on 1st Embodiment is performed. When each control signal of the ECU 33 is received, the brake actuator 20 and the hybrid system 21 perform the same operation as described in the automatic operation control device 1 according to the first embodiment.

  The automatic operation control device 3 has the same effects as the automatic operation control device 1 according to the first embodiment, and also has the following effects. According to the automatic driving control device 3, by learning the correction value of the maximum lateral acceleration and the deceleration corresponding to the road shape of the road entering after the curve of the preceding vehicle, the individual leading vehicle (individual driver) More accurate curve traveling characteristics according to the road shape after the curve can be obtained, and the speed pattern during the curve traveling of the preceding vehicle can be predicted with higher accuracy. As a result, a more optimal speed pattern of the host vehicle can be set, and effects such as improvement in fuel consumption of the host vehicle can be acquired to the maximum.

  As mentioned above, although embodiment which concerns on this invention was described, this invention is implemented in various forms, without being limited to the said embodiment.

  For example, in the present embodiment, it is applied to a hybrid vehicle that is automatically driven, but it can also be applied to a vehicle that performs various driving assistance (cooperative control) using a speed pattern for manual driving by a driver, The present invention is also applicable to a single drive source vehicle such as a motor.

  In the present embodiment, the preceding vehicle in front of the host vehicle is used as the surrounding vehicle. However, when there is another surrounding vehicle that affects the traveling of the host vehicle, the speed pattern of the surrounding vehicle is also predicted. The speed pattern of the surrounding vehicles may be taken into consideration.

  In the present embodiment, the speed pattern for generating fuel efficiency is generated. However, another travel plan such as an acceleration pattern may be generated, or a target speed pattern other than low fuel consumption may be used. .

  In the present embodiment, the ECU is configured by one ECU, but may be configured by a plurality of ECUs.

  In this embodiment, millimeter wave radar is applied to detect the preceding vehicle. However, other radar such as laser radar may be applied, or other means such as a stereo camera may be applied. Also good. Moreover, you may acquire the information of a preceding vehicle using a vehicle-to-vehicle communication.

  Further, in the present embodiment, the maximum lateral acceleration and deceleration at the time of curve driving of the preceding vehicle are learned, and the speed pattern at the time of curve driving of the preceding vehicle is obtained on the condition of the learned maximum lateral acceleration and deceleration at the time of curve driving. Although it is configured to predict, only one of the maximum lateral acceleration and deceleration may be learned, and the speed turn at the time of curve traveling of the preceding vehicle may be predicted on the condition of the learned one parameter.

  In the present embodiment, the maximum lateral acceleration and deceleration of the preceding vehicle are learned. However, standard values or the like may be used as the maximum lateral acceleration and deceleration of the preceding vehicle without learning.

  In the present embodiment, the maximum lateral acceleration is reduced when regenerating the speed pattern of the host vehicle. However, other conditions such as lowering the upper limit speed may be changed.

  Further, in the present embodiment, when the speed pattern of the own vehicle and the speed pattern of the preceding vehicle interfere with each other until the speed pattern of the own vehicle is regenerated in consideration of the speed pattern of the preceding vehicle, the corresponding vehicle Although the configuration is such that control is performed, if the ECU's calculation capability is high and the speed pattern of the host vehicle can be regenerated instantly, control corresponding to such interference is not necessary.

  In the third embodiment, the maximum lateral acceleration and deceleration correction values corresponding to the shape of the road after the curve (road width, etc.) are learned, and the learned maximum lateral acceleration and deceleration correction values are learned. The maximum lateral acceleration and deceleration are corrected according to the road shape of the destination after the curve, but learning may be performed based on the road shape of the curve itself. Thus, instead of correcting the maximum lateral acceleration and deceleration, the maximum lateral acceleration and deceleration according to the road shape may be learned during learning.

  In addition, learning about the lateral jerk in the second embodiment and learning about the correction value of the maximum lateral acceleration and the deceleration according to the road shape of the approach destination after the curve in the third embodiment are performed. A configuration may be adopted in which the speed pattern of the preceding vehicle is generated in consideration of both the learned lateral jerk and the correction value of the maximum lateral acceleration and deceleration corresponding to the road shape of the approach destination after the curve.

  DESCRIPTION OF SYMBOLS 1, 2, 3 ... Automatic operation control apparatus, 10 ... Wheel speed sensor, 11 ... Millimeter wave radar, 12 ... Navigation system, 20 ... Brake actuator, 21 ... Hybrid system, 22 ... Engine, 23 ... Motor, 24 ... Battery, 31, 32, 33 ... ECU, 31a, 32a, 33a ... preceding vehicle learning unit, 31b, 32b, 33b ... own vehicle speed pattern generation unit, 31c, 32c, 33c ... preceding vehicle speed pattern prediction unit, 31d, 32d, 33d ... Interference response unit, 31e, 32e, 33e ... Own vehicle speed pattern regeneration unit, 31f, 32f, 3f ... acceleration / deceleration control unit

Claims (4)

A vehicle control device that generates a travel plan of the host vehicle and performs vehicle control based on the travel plan,
A surrounding vehicle behavior predicting means for predicting a behavior of the surrounding vehicle at the time of curve driving based on a maximum lateral acceleration or / and a deceleration at the time of curve driving of the surrounding vehicle;
Travel plan generating means for generating a travel plan of the host vehicle based on the behavior at the time of curve driving of the surrounding vehicle predicted by the surrounding vehicle behavior predicting means ;
Learning means for learning the maximum lateral acceleration or / and deceleration at the time of the surrounding vehicle's curve driving when the surrounding vehicle travels the curve ,
The vehicle control apparatus according to claim 1, wherein the surrounding vehicle behavior predicting unit predicts a behavior of the surrounding vehicle during curve traveling based on the maximum lateral acceleration or / and the deceleration learned by the learning unit. The learning means learns the time differential value of the lateral acceleration when the surrounding vehicle travels the curve when the surrounding vehicle travels the curve,
2. The vehicle control apparatus according to claim 1 , wherein the surrounding vehicle behavior predicting unit predicts a behavior of the surrounding vehicle during curve driving based on a time differential value of the lateral acceleration learned by the learning unit. The learning means learns the maximum lateral acceleration or / and deceleration at the time of curve driving of the surrounding vehicle based on the road shape information of the curve when the surrounding vehicle travels the curve,
The surrounding vehicle behavior predicting means determines the behavior of the surrounding vehicle at the time of curve driving based on the maximum lateral acceleration or / and deceleration for each road shape of the curve learned by the learning means according to the road shape of the traveling curve. The vehicle control device according to claim 1 , wherein the vehicle control device predicts the vehicle. The travel plan for the host vehicle is a travel plan for the travel speed of the host vehicle,
The vehicle control device according to any one of claims 1 to 3, wherein the behavior of the surrounding vehicle during the curve traveling is a change in traveling speed during the curve traveling of the surrounding vehicle.

Images