JP5412985B2 - Vehicle travel support device - Google Patents

Description

  The present invention relates to a vehicle travel support apparatus that sequentially estimates a self-position of a host vehicle and uses the estimation result to assist the travel of the vehicle.

  Conventionally, a driving support device that guides a vehicle to a target position using automatic control or the like is known. In this type of device, the self-position is sequentially estimated through the amount of movement of the vehicle for each minute section, and various sensors for detecting the state quantity relating to the vehicle are mounted in order to perform the estimation. For example, there is a steering angle sensor that detects the steering angle of the front wheels through the rotation angle of the steering. However, when the vehicle loading weight changes or other vehicle characteristics change, the suspension geometry changes, so the wheel turning angle relative to the steering wheel angle may change.

  For example, in the method disclosed in Patent Document 1, in order to make the movement locus of the vehicle coincide with the true value even if the suspension geometry changes, it is determined whether or not the relationship between the movement distances of the left and right front wheels is a predetermined relationship. I'm researching. If the predetermined relationship is not satisfied, the vehicle can be correctly guided to the target position by correcting the movement locus according to the deviation. Patent Document 2 also discloses the concept of correcting sensor information deviation.

JP 2001-63599 A JP 2004-338637 A

  However, according to the methods disclosed in Patent Documents 1 and 2, the effect of the change of the steering angle with respect to the steering rotation angle due to the change of the suspension geometry and the change of other vehicle characteristics is analyzed in advance and the correction term is explicitly specified. Therefore, there is a problem that it is not possible to flexibly cope with variations in vehicle parameters for each vehicle.

  The present invention has been made in view of such circumstances, and an object thereof is to improve the estimation accuracy of the vehicle movement amount without directly analyzing the change in the vehicle characteristics.

In order to solve this problem, according to the present invention, the calculating means calculates the curvature of the traveling locus of the vehicle based on the vehicle wheel speed detected by the first detecting means and is detected by the second detecting means. The curvature is calculated based on the turning angle of the vehicle wheel . The evaluation unit evaluates the reliability of each curvature calculated by the calculation unit, and determines the curvature of the predetermined section based on the evaluation result. And an estimation means estimates the self position of the own vehicle based on the curvature of the predetermined area determined by the evaluation means.

  According to the present invention, it is possible to evaluate the reliability of the detection information of each detection means by comparing the same physical quantity (the amount of movement of the vehicle) calculated from the state quantities detected by the different detection means. it can. This makes it possible to use the amount of movement of the vehicle calculated from the individual detection information according to the reliability of the detection information, so there is no need to directly analyze changes in vehicle characteristics, The amount of movement can be determined with high accuracy. Therefore, the estimation accuracy can be improved by estimating the self-position based on the determined curvature.

Conceptual illustration of parking assistance Explanatory drawing explaining each parameter at the time of calculating the curvature of a vehicle locus from sensor information The block diagram which shows the structure of the vehicle travel assistance apparatus 1 The block diagram which shows the functional element which comprises the route calculating part 12. Flowchart showing the procedure of the sensor information reliability evaluation method The flowchart which shows the procedure of parking support processing

  Hereinafter, a vehicle travel support device according to an embodiment of the present invention will be described. This vehicle travel support device is a device that sequentially estimates the self position of the host vehicle and uses this estimation result to support the travel of the vehicle. The driving assistance in the present embodiment is assistance for guiding the vehicle based on the own position of the own vehicle so as to follow the target guidance route to the target position. For example, the driving assistance is guided to a predetermined parking position. Parking assistance. First, prior to description of a specific system configuration and operation of the vehicle travel support device, the concept of parking support according to the present embodiment will be briefly described with reference to FIG.

  The vehicle driving support device reversely parks the host vehicle relative to the initial position O at which parking is started, the relative target position (target parking position) P at which the vehicle is to be parked relative to the initial position O, and the target parking position P. Therefore, parking assistance is performed based on a target position (target return position) R that starts from the initial position O and turns the steering wheel. Specifically, the vehicle travel support device calculates a route (target guidance route) from the initial position O to the target switching position R, and guides the vehicle along the target guidance route. In addition, when the vehicle travel support device determines that the host vehicle has reached the target return position R, the vehicle travel support apparatus calculates a route (target guide route) from the target return position R to the target parking position P, and this target guide route Guide the vehicle along. The vehicle travel support device repeats the series of operations described above until it is determined that the target parking position P has been reached. In the present specification, the term “position” regarding the initial position O, the target parking position P, and the target return position R includes the posture (orientation) of the vehicle together with the position of the vehicle.

  When performing vehicle guidance, the vehicle travel support device sequentially estimates the self-position using sensor information obtained from sensors mounted on the vehicle, and performs feedback control based on the difference between the target guidance route and the estimated self-position. By performing, guidance to the target guidance route is performed. By the way, when performing such feedback control, it is necessary to accurately estimate the self-position, and therefore the reliability of the sensor information that is the premise is important.

  In the present embodiment, the reliability of individual sensor information is evaluated by comparing the same physical quantities calculated for each minute time with each other based on different types of sensor information. In the present embodiment, the curvature of the traveling locus of the vehicle is calculated as the same physical quantity, sensor information from the wheel speed sensors (left and right wheel speeds) is used as the first sensor information, and the second sensor information is used as the second sensor information. Sensor information (steering angle of the front wheels) from the steering angle sensor is used.

  FIG. 2 is an explanatory diagram for explaining each parameter when calculating the curvature of the vehicle trajectory from the sensor information. In the figure, (a) shows a case where the curvature ρst is obtained from the front wheel turning angle δ, and (b) shows a case where the curvature ρv is obtained from the left and right rear wheel speeds Vl and Vr. First, referring to (a), the curvature ρst calculated from the front wheel turning angle δ, which is sensor information from the steering angle sensor, is calculated based on the yaw rate ω (ω = dθ / dt) and the vehicle speed V (V = dL / dt). (L: mileage)), it can be expressed by the following formula.

数式１において、車両速度のＸ軸方向速度ｄx/ｄt、車両速度のＹ軸方向速度ｄy/ｄt、および、ｄθ/ｄtは、下式で示される。

In Equation 1, the X-axis direction speed dx / dt of the vehicle speed, the Y-axis direction speed dy / dt of the vehicle speed, and dθ / dt are expressed by the following expressions.

数式２において、Ｌｗは、車両のホイールベースである。数式１から分かるように、舵角センサのセンサ情報から得られる曲率ρstには、車輪速に関わる項がない。そのため、車輪速センサのセンサ情報に不確かさがあるようなシーンであっても、この曲率ρstの演算結果に直接的な影響が生じる虞がない。一方で、サスペンションジオメトリが変化した場合には、ステアリングホイール角に対する前輪の転舵角は変化する。そのため、このようなシーンでは、舵角センサのセンサ情報から得られる曲率ρstは不確かさを含むことがある。

In Equation 2, Lw is a vehicle wheelbase. As can be seen from Equation 1, the curvature ρst obtained from the sensor information of the rudder angle sensor has no term related to the wheel speed. For this reason, even in a scene where the sensor information of the wheel speed sensor is uncertain, there is no possibility that the calculation result of the curvature ρst will be directly affected. On the other hand, when the suspension geometry changes, the turning angle of the front wheels with respect to the steering wheel angle changes. Therefore, in such a scene, the curvature ρst obtained from the sensor information of the rudder angle sensor may include uncertainty.

  Further, referring to FIG. 4B, the curvature ρv calculated from the wheel speeds Vl and Vr of the left and right rear wheels obtained from the wheel speed sensor can be expressed by the following equation.

ここで、ｂは、後輪のトレッド長２ｂを表すパラメータである。また、車速Ｖ（後輪車軸の中心速度）は、（Ｖｒ＋Ｖｌ）／２であり、ヨーレートωは、（Ｖｒ−Ｖｌ）／２ｂである。数式３から分かるように、車輪速センサのセンサ情報から得られる曲率ρｖは、左右後輪の車輪速Ｖｌ，Ｖｒに依存する。車輪速は、車輪速センサが発生するパルス信号を用いて生成されるため、低速時には誤差が生じやすいという傾向を有している。そのため、低速シーンでは、車輪速センサのセンサ情報から得られる曲率ρｖは不確かさを含むことがある。

Here, b is a parameter representing the tread length 2b of the rear wheel. The vehicle speed V (center speed of the rear wheel axle) is (Vr + Vl) / 2, and the yaw rate ω is (Vr−Vl) / 2b. As can be seen from Equation 3, the curvature ρv obtained from the sensor information of the wheel speed sensor depends on the wheel speeds Vl and Vr of the left and right rear wheels. Since the wheel speed is generated using a pulse signal generated by the wheel speed sensor, an error tends to occur at a low speed. Therefore, in a low-speed scene, the curvature ρv obtained from the sensor information of the wheel speed sensor may include uncertainty.

  Therefore, in this embodiment, in order to evaluate the certainty of each sensor information, an evaluation function for evaluating the reliability of the curvatures ρst and ρv, which are the same physical quantities obtained from the sensor information, is defined. Then, the evaluation function is evaluated with a threshold value, thereby evaluating the probability of each sensor information. As described above, the sensor information (or curvature) evaluated to have high reliability is used for the sensor information (or curvature) used in the self-position estimation, thereby improving the self-position estimation accuracy. Can do.

  FIG. 3 is a block diagram illustrating a configuration of the vehicle travel support device 1 according to the present embodiment. The vehicle travel support device 1 is mainly configured by a parking support control unit 10, a detection system, and various actuators.

  The parking assistance control unit 10 can use a microcomputer mainly composed of a CPU, a ROM, a RAM, and an I / O interface. The parking assistance control unit 10 performs parking assistance for guiding the vehicle to a predetermined parking position by controlling various actuators based on various sensor information detected by the detection system. Information from various sensors and control units is input to the parking assist control unit 10.

  The parking support control unit 10 communicates with the brake control unit 20 that controls the brake state of the vehicle in an integrated manner, thereby detecting sensor information (the left and right rear wheels) detected by the brake control unit 20 through the sensors 20 and 21. Wheel speed, etc.). The wheel speed sensor 21 is provided on each of the left and right rear wheels, and each wheel speed sensor 21 detects the rotational speed of the wheel. For example, the wheel speed sensor 21 detects the rotation of a gear attached to the center of the wheel by a magnetic sensor, and outputs pulses corresponding to the rotation state in time series. The acceleration sensor 22 detects the acceleration in the front-rear direction. The brake control unit 20 can detect the wheel speeds of the left and right rear wheels based on the detection signals from the respective wheel speed sensors 21, and detects the longitudinal acceleration based on the detection signals from the acceleration sensor 22. be able to.

  The shift sensor 23 detects the position of the shift gear 43, that is, the shift position. The cameras 24 are installed at a plurality of locations around the vehicle, for example, at four locations on the front, rear, left and right. Each camera 24 images the surroundings of the vehicle and outputs the captured image to the parking assist control unit 10. As the input device 25, an operation panel including switches, a remote controller, a touch panel installed on the monitor 47, or the like can be used. The sonar sensors 26 are installed at a plurality of locations around the vehicle, and detect the direction and position of obstacles around the vehicle. For example, the sonar sensor 26 emits an ultrasonic wave from a transmitter that uses a piezoelectric element, and detects a reflected wave that is reflected when the ultrasonic wave hits an obstacle with a receiver that uses the same piezoelectric element. The communication device 27 can acquire information on a parking frame that can be parked in the parking lot by performing wireless communication with a communication device (not shown) provided as an infrastructure in the parking lot or the like. The steering angle sensor 28 is attached to the steering 46, and detects the steering angle of the front wheels corresponding to the rotation angle of the steering wheel.

  The parking assistance control unit 10 communicates with the brake control unit 20 to control the brake actuator 40 through the brake control unit 20, thereby controlling the operation of the brake 41 provided on the wheel. As the brake control unit 20, a microcomputer mainly composed of a CPU, a ROM, a RAM, and an I / O interface can be used. The brake actuator 40 is, for example, an actuator that controls a brake fluid pressure supplied to a wheel cylinder of the vehicle. By controlling the brake actuator 40, a braking force is generated on the wheels to perform a braking operation of the vehicle. Can do.

  Further, the parking assistance control unit 10 can change the position of the shift gear 43 by controlling the shift gear actuator 42. The parking assistance control unit 10 can communicate with the engine control unit 44. The engine control unit 44 is a controller for comprehensively performing electrical control in engine operation, and a microcomputer mainly composed of a CPU, a ROM, a RAM, and an I / O interface can be used. The parking assistance control unit 10 can control the state (for example, accelerator opening) of a drive source (not shown) such as an engine through the engine control unit 44.

  The parking assistance control unit 10 can adjust the rotation angle of the steering 46 by controlling the steering actuator 45. The steering actuator 45 is, for example, an electric actuator that applies an assist torque to the steering wheel. By controlling the steering actuator 45, the rotation angle of the steering 46, that is, an arbitrary turning angle is given to the wheel, and the vehicle is A steering operation can be performed. Furthermore, the parking assistance control unit 10 can display necessary information on the monitor 47 or notify the user via the speaker 48.

  The parking assistance control unit 10 includes a processing unit 11, a route calculation unit 12, a braking / driving control unit 13, and a steering control unit 14 when this is functionally grasped.

  The processing unit 11 recognizes (detects) the environment around the host vehicle by using one or more pieces of information obtained from the camera 24 or information obtained from the sonar sensor 26 or the like. The processing unit 11 detects the position and posture of the parking frame, the relative position of the obstacle, and the like based on the environment recognition, and recognizes the initial position O, the target parking position P, and the target return position R based on the detection result. And calculations related to the target guidance route. Further, the processing unit 11 calculates a control command for guiding the vehicle along the target guidance route based on the target guidance route and the self-position estimation result calculated by the route calculation unit 12.

  The route calculation unit 12 includes wheel speed sensor information V obtained from the wheel speed sensor 21, specifically, wheel speeds of the left and right rear wheels, and steering angle sensor information δ obtained from the steering angle sensor 28, specifically Is a so-called dead reckoning that obtains a travel locus of the vehicle based on the steered angle of the front wheels and the shift sensor information P obtained from the shift sensor 23, specifically, the shift position. (Self-position) is estimated sequentially.

  FIG. 4 is a block diagram showing functional elements constituting the route calculation unit 12. When this is functionally understood, the route calculation unit 12 has a travel distance calculation unit 12a, a first curvature calculation unit 12b, a second curvature calculation unit 12c, a reliability evaluation unit 12d, and a traveling direction determination. A unit 12e and a self-position estimating unit 12f.

  The travel distance calculation unit 12a calculates a travel distance (for example, the travel distance at the center of the rear wheel axle) based on the wheel speed sensor information V. The first curvature calculation unit 12b calculates the curvature ρst of the traveling locus of the vehicle based on the wheel speed sensor information V. The second curvature calculation unit 12c calculates the curvature ρv of the traveling locus of the vehicle based on the steering angle sensor information δ. Each curvature calculating unit 12b and 12c calculates curvatures ρst and ρv corresponding to a traveling locus within the minute time for each predetermined minute time. The reliability evaluation unit 12d evaluates the reliability of the wheel speed sensor information V and the steering angle sensor information δ based on the curvatures ρst and ρ calculated by the first and second curvature calculation units 12c, and this evaluation is performed. Based on the results, a sufficient curvature is determined. The determined curvature is output to the self-position estimation unit 12f. The traveling direction determination unit 12e determines the traveling direction of the vehicle based on the shift sensor information P. The self-position estimation unit 12f is a minute time based on the travel distance calculated by the travel distance calculation unit 12a, the traveling direction of the vehicle determined by the travel direction determination unit 12e, and the curvature determined by the reliability evaluation unit 12d. The travel locus of the vehicle at is obtained, and thereby the self position is estimated. The self-position can be estimated by numerically integrating the numerical formula 2 sequentially, but further includes information obtained by the camera 24 and the sonar sensor 26, that is, information such as a relative position of an obstacle around the own vehicle. The self position may be estimated.

  Referring again to FIG. 3, the braking / driving control unit 13 commands the accelerator opening to the engine control unit 44 so that the vehicle follows the target guidance route based on the control command from the processing unit 11. Further, the braking / driving control unit 13 gives a braking command to the brake control unit 20 when approaching the target switching position R or the target parking position P or when the sonar sensor 26 detects an obstacle. Further, the braking / driving control unit 13 controls the shift gear actuator 42 at a point where a shift change is necessary to automatically put the shift gear 43 into a desired gear.

  The steering control unit 14 monitors the steering angle of the front wheels detected by the steering angle sensor 28, and controls the steering actuator 45 so that the host vehicle follows the target guidance route based on the control command from the processing unit 11. To control.

  FIG. 5 is a flowchart illustrating a procedure of the sensor information reliability evaluation method according to the present embodiment. The process shown in this flowchart is called at a predetermined cycle during a parking assistance process described later, and is executed by the route calculation unit 12.

  First, in Step 1 (S1), the first curvature calculation unit 12b, as shown in Formula 3, is based on the wheel speeds Vl and Vr of the left and right rear wheels detected by the wheel speed sensor 21, and the vehicle in a minute time. The curvature ρv (hereinafter referred to as “curvature ρv based on wheel speed”) is calculated. On the other hand, in Step 2 (S2), the second curvature calculation unit 12c, as shown in Formula 1, based on the front wheel turning angle δ detected by the steering angle sensor 28, shows the vehicle travel locus in a minute time. The curvature ρst (hereinafter referred to as “curvature ρst based on the turning angle”) is calculated.

  In step 3 (S3), the reliability evaluation unit 12d determines the reliability regarding the curvatures ρst and ρv, which are the same physical quantities, in order to evaluate the reliability (probability) of the steering angle sensor information and the wheel speed sensor information. evaluate. In this embodiment, an evaluation function J (J = | ρv−ρst |) is defined. By this evaluation function J, the difference between the curvature ρv based on the wheel speed and the curvature ρst based on the turning angle is calculated as an evaluation criterion, and this evaluation criterion is evaluated with the threshold Th. When the evaluation criterion is equal to or greater than the threshold Th, the reliability evaluation unit 12d calculates that the curvature ρst based on the turning angle is higher than the curvature ρv based on the wheel speed. In other words, the rudder angle sensor information is evaluated as having higher reliability than the wheel speed sensor information. On the other hand, when the evaluation criterion is smaller than the threshold value Th, the reliability evaluation unit 12d calculates that the curvature ρv based on the wheel speed is higher than the curvature ρst based on the turning angle. In other words, it is evaluated that the wheel speed sensor information is more reliable than the steering angle sensor information. Here, it is preferable that the reliability evaluation unit 12d sets the above-described threshold Th based on the difference (| Δρst−Δρv |) of the variation (Δρst, Δρv) of the curvatures ρst, ρv for each minute time. .

  In step 4 (S4), the reliability evaluation unit 12d selects one sensor information with high reliability based on the evaluation result in step 3. Then, the reliability evaluation unit 12d determines the curvature calculated based on the selected sensor information as the final curvature, and outputs this to the self-position estimation unit 12f.

  FIG. 6 is a flowchart showing a procedure of parking support processing according to the present embodiment. The process shown in this flowchart is executed by the parking assistance control unit 10 using a parking assistance start instruction signal input through the operation of the input device 25 by the driver as a trigger. In addition, this parking support processing is performed by the driver until the host vehicle reaches the target parking position P (see FIG. 1) or until it is determined that the target parking position cannot be reached by one turnback trajectory. The parking assistance control unit 10 continues to execute until a parking assistance end instruction signal is input through the operation of the input device 25.

  First, in step 10 (S10), the processing unit 11 calculates a target parking position P with respect to the initial position O of the vehicle. Specifically, the processing unit 11 calculates (or designates) a target parking position P relative to the initial position O at which parking starts. For example, the processing unit 11 detects a white line indicating a parking frame, a parking wall, and the like using sensor information such as the camera 24 and the sonar sensor 26, and uses the detected information as a target parking position P as a target parking position P. calculate. In addition, when the processing unit 11 can communicate with the infrastructure facility via the communication device 27, when the information of the parking frame that can be parked can be obtained using the information from the infrastructure facility, the processing unit 11 includes this information. Based on this, the target parking position P may be calculated. Further, the processing unit 11 may display sensor information or information from infrastructure equipment on a monitor, and may cause the driver to specify the target parking position P using a display screen on the monitor.

  Then, the processing unit 11 determines whether or not parking is possible with respect to the target parking position P. When the initial position O is in a position relationship where parking is not possible with respect to the target parking position P, the processing unit 11 prompts the monitor 47 and the speaker 48 to change the initial position O or respecify the target parking position P. . On the other hand, when the initial position O is in a positional relationship where parking is possible with respect to the target parking position P, the processing unit 11 determines the calculated target parking position P as the final target parking position (step 11 (S11)).

  In step 12 (S12), the processing unit 11 calculates a target switching position R. Specifically, the processing unit 11 calculates a turnable area in a certain turning posture and presents it on the monitor 47. Here, when the driver selects an attitude angle at a desired turning point through the input device 25, the processing unit 11 causes the monitor 47 to present a turnable area in accordance with the driver's selection. When the driver arbitrarily selects one turning point from the turning possible area presented through the input device 25, the processing unit 11 sets the turning point selected by the driver as the target turning position R. Next, the processing unit 11 calculates a route (target guidance route) from the initial position O to the target return position R and a route (target guide route) from the target return position R to the target parking position P. To do.

  In step 12 (S12), the processing unit 11 presents the calculated target guidance route to the monitor 47, whereby the driver selects the target guidance route through the input device 25, whereby parking assistance is started. The On the other hand, when the driver does not select the target guidance route, the process returns to Steps 10 to 12, and the target parking position P or the target guidance route is set again.

  In Step 14 (S14), the route calculation unit 12 (specifically, the self-position estimation unit 12f) travels the vehicle calculated by the travel distance calculation unit 12a and the travel direction determination unit 12e. Based on the direction and the curvature calculated by the reliability evaluation unit 12d, the current self-position is estimated. The self-position is estimated by sequentially integrating numerical formula 2 into numerical values. Further, the processing unit 11 calculates a control command for guiding the vehicle along the target guidance route based on the target guidance route and the self-position estimation result calculated by the route calculation unit 12. Then, the braking / driving control unit 13 outputs a necessary control signal to the engine control unit 44, the brake control unit 20, or the shift gear actuator 42 based on the control command calculated by the processing unit 11. Further, the steering control unit 14 monitors the steering angle of the front wheels detected by the steering angle sensor 28, and sends a necessary control signal to the steering actuator 45 based on the control command calculated by the processing unit 11. Output. Thus, the vehicle is automatically controlled along the calculated target guidance route. When an obstacle that may come into contact with the vehicle is detected by detection information from the camera 24, the sonar sensor 26, or the like during automatic control of the vehicle, the processing unit 11 instructs the engine control unit 44 to decrease the accelerator opening. And a braking command is given to the brake control unit 20. The processing unit 11 always performs dead reckoning during traveling by automatic control, feeds back an error from the target guidance route, and performs control so that the difference between the target guidance route and the actual traveling route is minimized.

  In step S15 (S15), the processing unit 11 determines whether or not an abnormality is detected such that the self-position is greatly deviated from the target guidance route. If an abnormality exists, step 19 (S19) described later is performed. Proceed to the process. On the other hand, if no abnormality exists, the processing unit 11 proceeds to the process of step 16 (S16).

  In step 16, the processing unit 11 determines whether or not the host vehicle has reached the target switching position R and a gear change is necessary. If an affirmative determination is made in step 16, that is, if a gear change is necessary, the processing unit 11 gives a braking command to the brake control unit 20 through the braking / driving control unit 13 to stop the vehicle and shift gear actuator 42. To shift the shift gear 43 to the desired range, and then proceed to Step 17. On the other hand, if a negative determination is made in step 16, that is, if a gear change is not necessary, the process returns to step 14.

  In step 17 (S17), the processing unit 11 determines whether or not the host vehicle has reached the target parking position P. If the determination in step 17 is affirmative, that is, if the host vehicle has reached the target parking position P, the processing unit 11 gives a braking command to the brake control unit 20 through the braking / driving control unit 13 to stop the vehicle. In both cases, the shift gear actuator 42 is driven to change the shift gear 43 to the parking range, and then the process proceeds to step 18 (S18). In step 18, the processing unit 11 presents on the monitor 47 that the parking support is to be ended, and ends the driving support. On the other hand, if a negative determination is made in step 17, that is, if the host vehicle has not reached the target parking position P, the process returns to step 14.

  In step 19, the processing unit 11 notifies the driver of the abnormality via the monitor 47 or the speaker 48. In step 20, the processing unit 11 determines whether it is possible to return to the route leading to the target parking position P. If an affirmative determination is made in step 20, that is, if it is possible to return to the route, the process returns to step 12. On the other hand, if a negative determination is made in step 20, that is, if it is impossible to return to the route, the process proceeds to step 21 (S21). Then, in step 21, the processing unit 11 presents that the parking assistance is stopped on the monitor 47, and ends the traveling assistance.

  As described above, according to the present embodiment, the curvature ρv based on the wheel speed and the curvature ρst based on the steering angle, which are the same physical quantities, are compared with each other, so that the wheel speed sensor information or the steering angle sensor information can be compared. Reliability can be evaluated. Depending on the reliability of the sensor information, the curvatures ρst and ρv calculated from the sensor information can be used, so that the curvature can be determined more accurately than when each sensor information is used alone. . Further, by using a plurality of sensor information according to the reliability, these can be used in a mutually complementary relationship, so there is no need to directly analyze changes in vehicle characteristics. Therefore, the estimation accuracy can be improved by estimating the self-position based on the determined curvature.

  Further, according to the present embodiment, the reliability of the sensor information can be easily evaluated by comparing the evaluation standard with the threshold Th using the difference between the curvatures ρst and ρv obtained from the sensor information as the evaluation standard. . The threshold Th may be set based on the difference between the amounts of change in the curvatures ρst and ρv for each predetermined section, and is obtained by simulation and experiment. However, the threshold Th is not a multidimensional vector amount, but a scalar amount serving as the threshold Th. Therefore, it is easy to determine the threshold value.

  The vehicle travel support apparatus according to the present invention is not limited to the one described in the above-described embodiment, and various modifications can be made within the scope of the invention. For example, the evaluation of the reliability of the curvatures ρst and ρv is not limited to the evaluation function J described above, but is a statistic indicating the reliability of the sensor information obtained by each detection means (the wheel speed sensor 21 and the steering angle sensor 28). You may evaluate based on. According to such a method, it is possible to evaluate the reliability not only for switching two types of sensor information but also for arbitrary plural pieces of sensor information, so that the curvature can be accurately determined. Therefore, the estimation accuracy can be improved by estimating the self-position based on the determined curvature.

  In the above-described embodiment, the curvature of the curvatures ρv and ρst that is evaluated as having high reliability is determined as the curvature of minute time. However, in addition to this, depending on the reliability of each curvature. The curvature based on the curvatures ρv and ρst may be determined by using a weighted average obtained by weighting. Thereby, a curvature can be determined more accurately than the case where sensor information is used alone. Therefore, the estimation accuracy can be improved by estimating the self-position based on the determined curvature.

  In the above-described embodiment, the method of automatically controlling the own vehicle has been described. However, the present invention can also be applied to driving assistance in which the driver operates the steering wheel along the instructed route. For example, the rotation angle of the steering 46 at the initial position O, the target return position R and the rotation angle of the steering 46 from the target return position R to the target parking position P are designated, and the steering 46, brake 41 and shift are driven. It is like having the hand operate.

  Moreover, in embodiment mentioned above, it is the method of estimating a self-position by calculating a curvature for every minute time. However, the unit of curvature calculation is not limited to time, and may be calculated for each minute distance. In this case, each mathematical formula described above may be a differential formula related to the travel distance. As described above, in the present invention, the curvature calculated based on the sensor information is defined based on the distance even if it is defined based on the time of the section as long as it is calculated for each predetermined section. It may be.

  In the above-described embodiment, the curvature of the travel locus is given as the physical quantity to be compared. However, the present invention is not limited to this, and the present invention is applicable to various parameters that include the curvature of the travel locus and represent the amount of movement of the vehicle. Can do.

DESCRIPTION OF SYMBOLS 1 ... Vehicle driving assistance apparatus 10 ... Parking assistance control unit 11 ... Processing part 12 ... Path | route calculating part 12a ... Travel distance calculating part 12b ... 1st curvature calculating part 12c ... 2nd curvature calculating part 12d ... Reliability evaluation part 12e ... running direction determination unit 12f ... self-position estimation unit 13 ... brake drive control unit 14 ... steering control unit 20 ... brake control unit 21 ... wheel speed sensor 22 ... acceleration sensor 23 ... shift sensor 24 ... camera 25 ... input device 26 ... sonar sensor DESCRIPTION OF SYMBOLS 27 ... Communication apparatus 28 ... Steering angle sensor 40 ... Brake actuator 41 ... Brake 42 ... Shift gear actuator 43 ... Shift gear 44 ... Engine control unit 45 ... Steering actuator 46 ... Steering 47 ... Monitor 48 ... Speaker

Claims (7)

In the vehicle travel support device that sequentially estimates the self position of the host vehicle and uses the estimation result to support the travel of the vehicle,
First detecting means for detecting wheel speed of the vehicle;
Second detecting means for detecting a turning angle of the wheel of the vehicle;
Calculating means for calculating the curvature of the traveling locus of the vehicle based on the wheel speed for each predetermined section and calculating the curvature based on the turning angle of the wheel ;
Evaluation means for evaluating the reliability of each curvature calculated by the calculation means, and determining the curvature of the predetermined section based on the evaluation result;
A vehicle travel support apparatus comprising: an estimation unit configured to estimate a self-position of the host vehicle based on the curvature of the predetermined section determined by the evaluation unit. The evaluation means calculates a difference between the curvature based on the wheel speed and the curvature based on the turning angle of the wheel as an evaluation criterion, and based on the turning angle of the wheel when the evaluation criterion is equal to or greater than a threshold value. When the curvature is calculated to be more reliable than the curvature based on the wheel speed, and the evaluation criterion is smaller than the threshold, the curvature based on the wheel speed is based on the turning angle of the wheel. The vehicle travel support apparatus according to claim 1 , wherein the vehicle travel support apparatus calculates that the reliability is higher than that. 3. The vehicle travel support apparatus according to claim 2 , wherein the evaluation unit sets the threshold value based on a difference in a change amount of each curvature for each predetermined section. 4. The evaluation means evaluates the reliability of the curvature based on the wheel speed and the curvature based on the turning angle of the wheel based on a statistic indicating the reliability of detection information obtained by each detection means. The vehicle travel support apparatus according to claim 1 , wherein 2. The vehicle travel support apparatus according to claim 1 , wherein the evaluation unit determines, as a curvature of the predetermined section, a curvature that is evaluated as having a high reliability among the curvatures. 2. The vehicle according to claim 1 , wherein the evaluation unit determines the curvature of the predetermined section based on each curvature by using a weighted average obtained by weighting according to the reliability of each curvature. Driving support device. An obstacle detection means for detecting a relative position of the obstacle around the host vehicle;
Said evaluation means, and the curvature based on the wheel speed, and the curvature based on the steering angle of the wheel, on the basis of the detection result of the obstacle detecting means, and determining the curvature of the predetermined section The vehicle travel support apparatus according to claim 1.

Images