Vehicle surrounding moving object detection device

Abstract

To provide a moving object detecting device around a vehicle, which is capable of preventing an erroneous detection even for an object of a different type of detection target such as a vehicle and a cyclist in a collision determination between the own vehicle and the cyclist or the like.SOLUTION: A collision detection emergency call device comprising a moving object detecting device around a vehicle, can perform collision determination to prevent a false detection depending on a situation, since: a collision possibility is calculated by comparing a predicted traveling route of the own vehicle calculated based on driving information of the own vehicle with a predicted traveling route of a cyclist calculated based on the peripheral object information; a determination threshold value to determine a presence or absence of the collision is changed based on the calculated collision possibility; and the collision with the cyclist is detected based on the changed determination threshold value.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a vehicle periphery moving object detection device. [Background technology]

[0002] Conventionally, vehicles are provided with collision detection devices for deploying airbags inside or outside the vehicle, etc. Also, a protection control device has been proposed that detects the relative position (height of the collision position) of a pedestrian or a rider on a two-wheeled vehicle (hereinafter collectively referred to as a cyclist) moving around the vehicle, and controls an airbag outside the vehicle, etc. (see Patent Document 1).

[0003] This protection control device changes the deployment angle θ of the main airbag provided in the front of the vehicle using a sub-airbag based on the height of the collision position and the estimated height of the center of gravity of the obstacle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2011-218857 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, there are cases where a cyclist may collide with something other than the front of the vehicle, such as hitting the side of the vehicle when turning right or left. In addition, there is a risk that such a vehicle may not be able to provide sufficient protection for the cyclist if the cyclist collides with the side of the vehicle. In addition, unlike the front of the vehicle, there are many obstacles on the sides of the vehicle, and even if the moving object detection device of the protection and control device detects a cyclist, the cyclist is often subsequently blocked by an obstacle and lost sight of.

[0006] Furthermore, existing passenger protection airbag sensors and the like can be considered as a means for detecting a cyclist's collision with the side or rear of a vehicle. In these passenger protection airbag sensors, a detection threshold is used to determine whether or not a collision with another vehicle or the like has occurred. However, the detection threshold of existing passenger protection airbag sensors is designed to detect a collision with a vehicle or the like, and is too high to detect a collision of a cyclist. For this reason, in order to be able to detect a cyclist, the detection threshold of existing passenger protection airbag sensors must be kept low. However, if the detection threshold of existing passenger protection airbag sensors is made too low, they will erroneously detect a collision even when it is not necessary.

[0007] The present invention has been made to solve such conventional problems, and has as its objective the provision of a vehicle surrounding moving object detection device that can prevent erroneous detection even when detecting different types of objects, such as vehicles and cyclists. [Means for solving the problem]

[0008] The vehicle surrounding moving object detection device according to the present invention comprises: a host vehicle driving information acquisition means for acquiring driving information of the host vehicle; a collision detection means for detecting a collision between the host vehicle and an external object by changing a detection level based on a predetermined judgment threshold; a surrounding object information acquisition means for acquiring object information around the driving position of the host vehicle; a travel path prediction control unit for predicting a travel path of the host vehicle based on the travel information of the host vehicle, and predicting a travel path of a pedestrian or cyclist around the host vehicle based on the object information; and a collision possibility calculation control unit for calculating a collision possibility by comparing the predicted travel path of the host vehicle predicted by the travel path prediction control unit with the predicted travel path of the pedestrian or cyclist. before The collision detection means determines whether or not a collision occurs by setting the determination threshold value. setting a determination threshold value change control unit for changing a threshold value of the surroundings of the vehicle; Includes at least one of intersections and slopes a surrounding geographic information acquisition control unit that acquires geographic information, the determination threshold change control unit sets a determination threshold of a lower value as the collision possibility increases,The travel path prediction control unit is characterized in that when the pedestrian or cyclist that was captured by the surrounding object information acquisition means can no longer be captured, the travel path prediction control unit corrects the predicted travel path of the pedestrian or cyclist based on the geographic information acquired by the surrounding geographic information acquisition control unit.

[0009] In the vehicle surrounding moving object detection device according to the present invention, the object information used by the travel path prediction control unit to predict the travel path includes the pedestrian or the cyclist. Velocity vector from detection to the present The information may be included.

[0010] Furthermore, in the vehicle surrounding moving object detection device according to the present invention, the collision possibility calculation control unit calculates a predicted travel path of the host vehicle and Pedestrians or cyclists and a predicted collision time based on the predicted travel path of the vehicle, thereby calculating the possibility of collision. Effect of the Invention

[0011] According to the present invention, it is possible to provide a vehicle surrounding moving object detection device that can prevent erroneous detection even when detecting different types of objects, such as a vehicle and a cyclist. [Brief description of the drawings]

[0012] [Figure 1] 1 is a block diagram showing an outline of a collision detection and emergency notification device including a vehicle surrounding moving object detection device according to an embodiment of the present invention; [Diagram 2] 5 is a flowchart showing a collision risk determination process of the collision detection and emergency notification device according to the embodiment of the present invention. [Diagram 3] 13 is a flowchart showing a capture-time collision predicted time calculation process of the collision risk determination process. [Figure 4] 10 is a flowchart showing a collision prediction time estimation process of the collision risk determination process. [Diagram 5] FIG. 13 is a conceptual diagram of a decision threshold value. [Figure 6] FIG. 1 is a conceptual diagram of pre-vector prediction. [Figure 7] This is a conceptual diagram of a cyclist's predicted vector and a vehicle's predicted vector, as well as a prediction concept of the predicted time to collision (TTC). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In this embodiment, a collision detection and emergency notification device that is equipped in a vehicle and uses a vehicle surrounding moving object detection device will be described.

[0014] Fig. 1 is a block diagram showing an outline of a collision detection emergency call device including a vehicle surrounding moving object detection device in an embodiment of the present invention. Fig. 2 is a flowchart showing a collision risk determination process of the collision detection emergency call device in an embodiment of the present invention. Fig. 3 is a flowchart showing a capture-time collision prediction time calculation process of the collision risk determination process. Fig. 4 is a flowchart showing a collision prediction time estimation process of the collision risk determination process.

[0015] (Configuration of collision detection emergency notification device 1) As shown in FIG. 1, the collision detection emergency call device 1 includes an object detection radar 10, a collision detection sensor 20, a GPS 31, a gyro sensor 32, a VDC 33, an EPS 34, a turn indicator 35, a control ECU 100, and an emergency call control ECU 200.

[0016] (Object Detection Radar 10) The object detection radar 10 acquires information about objects in the vicinity of the vehicle's traveling position (hereinafter referred to as surrounding object information), for example, detecting the relative distance and relative speed between cyclists around the vehicle and the vehicle (the vehicle itself). The object detection radar 10 emits radio waves to detect objects such as obstacles and cyclists, and measures the distance and direction to the objects. The object detection radar 10 is attached to the front bumper or rear bumper of the vehicle, and is capable of monitoring the front, sides, rear, etc. of the vehicle. An example of the positions of the vehicle and the cyclist is shown in Figure 7. The object detection radar 10 also includes a left front corner radar 11, a right front corner radar 12, a left rear corner radar 13, and a right rear corner radar 14.

[0017] The left front corner radar 11 is provided on the left side of the front (forward) of the vehicle, and detects surrounding objects in front and to the left of the vehicle, with the left front being the center. The right front corner radar 12 is provided on the right side of the front (forward) of the vehicle, and detects surrounding objects in front and to the right of the vehicle, with the right front being the center.

[0018] The left rear corner radar 13 is provided on the left side of the rear (rear) of the vehicle, and detects surrounding objects to the rear and left of the vehicle, with the left rear area being the center. The right rear corner radar 14 is provided on the right side of the rear (rear) of the vehicle, and detects surrounding objects to the rear and right of the vehicle, with the right rear being the center.

[0019] (Collision detection sensor 20) The collision detection sensor 20 detects a collision with a surrounding object. For example, the collision detection sensor 20 detects the collision acceleration (collision G) when a surrounding object collides with the vehicle, the pressure when the surrounding object collides with the vehicle, etc. The collision detection sensor 20 includes a collision G sensor that detects collision G, a pressure sensor that detects pressure information, etc. The collision detection sensor 20 may also include a vehicle G sensor that obtains information for predicting a collision in advance, for example, a vehicle G sensor that detects acceleration / deceleration of the vehicle.

[0020] (GPS31) The GPS 31 acquires GPS information, and the GPS information includes the vehicle's position. The GPS 31 receives signals from several satellites in the sky to detect the vehicle's position. As described later, the control ECU 100 acquires surrounding geographic information of the vehicle's position from the information server 300 based on the vehicle's position acquired by the GPS 31. The surrounding geographic information of the vehicle's position may be stored in advance in a storage area of ​​the control ECU 100, and the surrounding geographic information may be acquired from the vehicle's position information in the acquired GPS information.

[0021] (Gyro sensor 32) The gyro sensor 32 acquires the inclination of the vehicle by detecting the angular velocity and detecting inclination information of the vehicle.

[0022] (VDC33) The VDC 33 acquires the host vehicle speed. The VDC 33 prevents the vehicle from skidding and improves driving stability during cornering, etc., and in this example, the host vehicle speed detected by the VDC 33 is acquired.

[0023] (EPS34) The EPS 34 acquires the steering angle. The EPS 34 is an electric power steering that assists the steering force by electricity (motor), and here, the steering angle (steering angle) detected by the EPS 34 is acquired.

[0024] (turn signal 35) The direction indicator 35 is a device for notifying the surroundings of the traveling direction of the vehicle when the vehicle turns right or left or changes course.

[0025] (Control ECU 100) The control ECU 100 detects a moving object around the vehicle, determines whether or not there is a collision with a surrounding object, and sends an emergency call trigger to the emergency call control ECU 200 based on the result of the determination. The control ECU 100 also includes a CPU, a ROM, a RAM, an EEPROM consisting of a rewritable non-volatile memory, and an input / output interface circuit, and is configured to centrally control the collision detection and emergency notification device 1.

[0026] The control ECU 100 is also connected to the object detection radar 10, the collision detection sensor 20, the GPS 31, the gyro sensor 32, the VDC 33, the EPS 34, the direction indicator 35, and the emergency call control ECU 200 via an input / output interface circuit.

[0027] The control ECU 100 then acquires surrounding object information from the object detection radar 10, and predicts and estimates a collision with a surrounding object based on the vehicle position acquired from GPS 31, the vehicle speed acquired from VDC 33, the steering angle acquired from EPS 34, the turn indicator state acquired from turn indicator 35, etc., and outputs an emergency call trigger to the emergency call control ECU 200 if there is a risk of collision with a surrounding object.

[0028] (Emergency call control ECU 200) The emergency call control ECU 200 transmits an emergency call signal to the network when a risk of collision with a surrounding object occurs. That is, the emergency call control ECU 200 transmits the emergency call signal to the network when an emergency call trigger is received from the control ECU. Note that, when a risk of collision with a surrounding object occurs, the emergency call control ECU 200 may also notify the driver (passenger) of the occurrence of the collision risk.

[0029] (Operation of collision detection emergency notification device 1) Next, the operation of the collision detection and emergency notification device 1 will be described. 2 to 4 are flowcharts showing a collision risk calculation process using a vehicle surrounding moving object detection process in the collision detection and emergency notification device 1. FIG.

[0030] The collision detection emergency notification device 1 has a control ECU 100, which periodically performs the collision risk calculation process shown in Figures 2 to 5 at predetermined intervals. Note that this collision risk calculation process may be performed based on the input of a predetermined detection signal. In the collision risk calculation process, the control ECU 100 first performs a host vehicle information acquisition process (step S11).

[0031] Next, the control ECU 100 determines whether or not the vehicle speed of the host vehicle exceeds 0 [km / h] (step S12). That is, the control ECU 100 determines whether or not the vehicle is moving forward. In this determination process, it may be determined whether the vehicle speed is not 0 [km / h]. In this case, the control ECU 100 determines whether the vehicle is stopped, and performs the following threshold value change process even if the vehicle speed is less than 0 [km / h], that is, when the vehicle is backing up (moving backward).

[0032] If the control ECU 100 determines that the vehicle speed exceeds 0 [km / h] (determines YES in step S12), it proceeds to step S13, and if it determines that the vehicle speed does not exceed 0 [km / h] (determines NO in step S12), it returns to step S11.

[0033] When the control ECU 100 determines that the vehicle speed exceeds 0 [km / h] (determined as YES in step S12), the control ECU 100 performs a surrounding object information acquisition process (step S13). Specifically, the control ECU 100 acquires surrounding object information detected by the object detection radar 10.

[0034] Next, the control ECU 100 determines whether or not a cyclist has been detected around the vehicle based on the acquired peripheral object information (step S14). At this stage, the control ECU 100 only determines whether a cyclist has been provisionally detected. That is, the control ECU 100 detects a moving object in the vicinity based on the relative speed between the vehicle speed and the peripheral objects, and if a moving object is present in the vicinity, it provisionally detects the moving object as a cyclist. In this embodiment, a cyclist includes all types of people, including pedestrians, bicycle riders, and motorcycle riders.

[0035] If the control ECU 100 determines that a cyclist has been detected (YES in step S14), it proceeds to step S15, and if it determines that a cyclist has not been detected (NO in step S14), it returns to step S11.

[0036] When the control ECU 100 determines that a cyclist has been detected (YES in step S14), it determines whether the relative speed V1 between the vehicle and the cyclist is 30 km / h or more (step S15). If the relative speed V1 between the vehicle and the cyclist is less than 30 km / h, it is possible that the cyclist is traveling alongside the vehicle, and it can be determined that there is a high risk of collision.

[0037] If the control ECU 100 determines that the relative speed V1 is 30 km / h or greater (determines YES in step S15), it proceeds to step S16, and if it determines that the relative speed V1 is not 30 km / h or greater (determines NO in step S15), that is, that the relative speed V1 is less than 30 km / h, it proceeds to step S17.

[0038] When the control ECU 100 determines that the relative speed V1 is 30 km / h or more (determined as YES in step S15), it sets the side contact monitoring threshold value "1" as the target threshold value flag (step S16). Note that the target threshold value flag is set to the threshold value "0" in the initial state (at the start of the collision risk calculation process). Here, the target threshold flag determines which of a plurality of thresholds is to be used when the collision detection sensor 20 detects a collision with an external object, and determines the detection level for determining a collision. That is, in this embodiment, there are "four levels" of judgment thresholds, and the target threshold flag determines which threshold is to be used when determining a collision.

[0039] A conceptual diagram of the judgment threshold that changes depending on multiple side contact monitor thresholds is shown in Figure 5. As shown in Figure 5, the judgment threshold when the sensor or the like detects a collision changes in stages depending on the side contact monitor threshold that is set.

[0040] Specifically, the side contact monitoring judgment threshold has four values, starting with the initial threshold value "0", followed by threshold value "1", threshold value "2", and threshold value "3". The judgment threshold value for determining whether or not the value detected by the collision detection sensor 20 is to be judged as a collision changes depending on the set side contact monitoring judgment threshold value. Therefore, as the number increases from threshold value "1" to threshold value "2", the judgment threshold value decreases, and even a smaller impact is judged as a collision, with threshold value "3" being the most sensitive to judging a collision. When the relative speed V1 is 30 km / h or higher, the risk of collision with a cyclist is considered to be low, but since a cyclist is detected near the vehicle, the side contact monitoring threshold is set to "1", which is "one step" higher (the judgment threshold is smaller) than the initial side contact monitoring threshold "0", and the sensitivity is set higher than in the initial state. In other words, the collision detection sensor 20 will detect a collision with an external object even with a smaller detection value than in the initial state.

[0041] On the other hand, when the control ECU 100 determines that the relative speed V1 is less than 30 [km / h] (NO in step S15), it sets the side contact monitoring threshold value "2" as the target threshold value flag (step S17). In other words, when the relative speed V1 is less than 30 [km / h], it is possible that the vehicle is traveling alongside a cyclist, and the risk of collision is higher than when the relative speed V1 is 30 [km / h] or more. Therefore, the side contact monitoring threshold value is set to "2", one level higher than the side contact monitoring threshold value "1" set when the relative speed V1 is 30 [km / h] or more, to increase the sensitivity.

[0042] After setting the side contact monitoring threshold value (step S16 or step S17), the control ECU 100 determines whether or not the cyclist has been lost (step S18), that is, whether or not the cyclist that has already been detected has been detected without losing sight of him or her.

[0043] If the control ECU 100 determines that the cyclist has been lost (determined as YES in step S18), it proceeds to step S19, and if it determines that the cyclist has not been lost (determined as NO in step S18), it proceeds to step S21.

[0044] When the control ECU 100 determines that the cyclist has been lost (determined as YES in step S18), it determines whether or not there is an obstacle to the side (step S19). Here, the control ECU 100 can recognize, for example, a stopped nearby object as an obstacle. The control ECU 100 can determine whether or not the nearby object is stopped based on the vehicle speed of the vehicle and the relative speed between the vehicle and the nearby object.

[0045] If the control ECU 100 determines that there is an obstacle on the side (determined as YES in step S19), it shifts the process to step S22, and if it determines that there is no obstacle on the side (determined as NO in step S19), it shifts the process to step S20.

[0046] If the control ECU 100 determines that there is no obstacle on the side (NO in step S19), it redetects the cyclist and determines whether or not the cyclist has been detected again (step S20).

[0047] If the control ECU 100 determines that the cyclist has been redetected (YES in step S20), it proceeds to step S21, and if it determines that the cyclist cannot be redetected (NO in step S20), it proceeds to step S22.

[0048] When the control ECU 100 determines that the cyclist has not been lost (NO in step S18) or has redetected the cyclist (YES in step S20), it performs a capture-time collision prediction time calculation process (step S21). In this collision prediction time calculation process, the travel path of each cyclist is predicted from the operation information of the vehicle and the information of the captured cyclist, the predicted travel paths are compared, the collision possibility is calculated, and the collision prediction time (TTC) is calculated. For example, signs of right or left turns are evaluated from vehicle signals (the state of the turn indicator 35, the steering angle), and if there is a sign that the vehicle's travel direction is changing, it is determined that the risk of contact with the cyclist is increasing, and a low threshold value is set to make it easier to detect a side collision. Details of the collision prediction time calculation process when captured will be described later.

[0049] On the other hand, if the control ECU 100 determines that there is an obstacle on the side (YES in step S19) or determines that the cyclist cannot be detected again (NO in step S20), it performs a collision prediction time estimation process (step S22). In this collision prediction time estimation process, the cyclist's travel route is predicted based on the riding conditions of the cyclist immediately before he / she was lost, and the predicted travel route of the cyclist is corrected based on surrounding geographic information. The vehicle's travel route is also predicted based on the vehicle's operation information (steering angle, etc.) and vehicle speed. The cyclist's predicted travel route is then compared with the vehicle's predicted travel route to calculate the possibility of collision and estimate the predicted time to collision (TTC). Details of the collision prediction time estimation process will be described later.

[0050] When the control ECU 100 finishes the capture-time collision prediction time calculation process (step S21) or the collision prediction time estimation process (step S22), it determines whether the calculated collision prediction time (TTC) is equal to or less than a predetermined first determination value (step S23). Here, the first determination value is a predetermined time for determining whether the time until the collision prediction between the cyclist and the vehicle is short and the collision risk is high. The control ECU 100 determines whether or not a collision has occurred based on the collision prediction time (TTC) calculated in the process of step S23 and the process of step S24. That is, if the collision prediction time (TTC) is short, it is determined that it is difficult to avoid a collision, and a threshold value that makes it easier to detect a collision is set.

[0051] If the control ECU 100 determines that the collision prediction time (TTC) is less than or equal to the first judgment value (YES in step S23), it transfers processing to step S25, and if the control ECU 100 determines that the collision prediction time (TTC) is not less than or equal to the first judgment value (NO in step S23), it transfers processing to step S24.

[0052] When the control ECU 100 determines that the collision prediction time (TTC) is not equal to or less than the first judgment value (NO in step S23), it determines whether the collision prediction time (TTC) is equal to or greater than a predetermined second judgment value (step S24). Here, the second judgment value is a predetermined time for determining whether the time until the collision prediction between the cyclist and the vehicle is long and there is no collision risk or is very small. Therefore, the second judgment value is a larger value than the first judgment value, i.e., a longer time.

[0053] If the control ECU 100 determines that the collision prediction time (TTC) is greater than or equal to the second judgment value (YES in step S24), it transfers processing to step S27, and if the control ECU 100 determines that the collision prediction time (TTC) is not greater than or equal to the second judgment value (NO in step S24), it transfers processing to step S26.

[0054] If the control ECU 100 determines that the collision prediction time (TTC) is less than or equal to the first judgment value (YES in step S23), it sets the side contact monitoring threshold value "3" as the target threshold value flag (step S25) and terminates the collision risk calculation process. In other words, when the predicted time to collision (TTC) is equal to or less than the first judgment value, it is determined that there is a high risk of collision with a cyclist, and the highest (smallest) side contact monitoring threshold value is set to "3", thereby setting the sensitivity to the highest level.

[0055] On the other hand, if the control ECU 100 determines that the collision prediction time (TTC) is not greater than or equal to the second judgment value, that is, that the collision prediction time (TTC) is greater than or equal to the first judgment value and less than the second judgment value (NO in step S24), it retains the side contact monitoring threshold value that has already been set (step S26) and terminates the collision risk calculation process. In other words, if the collision prediction time (TTC) is greater than or equal to the first judgment value and less than the second judgment value, the risk of collision with a cyclist is determined to be medium to low, and the side contact monitoring threshold value that has already been set according to the collision risk is set.

[0056] In addition, if the control ECU 100 determines that the collision prediction time (TTC) is greater than or equal to the second judgment value (determined as YES in step S24), it sets the side contact monitoring threshold value "0" as the target threshold value flag (step S27) and terminates the collision risk calculation process. In other words, if the collision prediction time (TTC) is equal to or greater than the second judgment value, it is determined that there is no risk of a collision with a cyclist or that the risk is extremely low, and the side contact monitoring threshold is set to "0" and the judgment threshold is restored to the threshold for normal riding.

[0057] (Collision prediction time calculation process at capture time) Next, the capture-time collision predicted time calculation process performed in step S21 of Fig. 2 will be described with reference to Fig. 3. Fig. 3 is a diagram showing a subroutine of the capture-time collision predicted time calculation process.

[0058] In the capture-time collision prediction time calculation process, the control ECU 100 first determines whether the direction indicator 35 is ON (step S31). That is, the control ECU 100 determines whether the state of the direction indicator 35 is the left ON state, the right ON state, or both the left and right OFF states. If the control ECU 100 determines that the direction indicator 35 is ON (determines YES in step S31), it proceeds to step S32, and if it determines that the direction indicator 35 is not ON (determines NO in step S31), it proceeds to step S35.

[0059] When the control ECU 100 determines that the direction indicator 35 is ON (determined as YES in step S31), the control ECU 100 determines whether the steering angle is equal to or larger than a predetermined determination angle (step S32). That is, the control ECU 100 determines whether the steering angle acquired from the EPS 34 is equal to or larger than a predetermined determination angle.

[0060] If the control ECU 100 determines that the steering angle is equal to or greater than the predetermined judgment angle (determines YES in step S32), it proceeds to step S33, and if it determines that the steering angle is not equal to or greater than the predetermined judgment angle (determines NO in step S32), it proceeds to step S34.

[0061] When the control ECU 100 determines that the steering angle is equal to or greater than the predetermined determination angle (determined as YES in step S32), it sets the side contact monitoring threshold value "3" as the target threshold value flag (step S33) and proceeds to step S36. In other words, when the steering angle is equal to or greater than a predetermined judgment angle, it is determined that there is a high risk of collision with a cyclist, and the side contact monitoring threshold value is set to the highest (smallest value) of "3," thereby setting the sensitivity to the highest level.

[0062] On the other hand, if the steering angle is not equal to or greater than the predetermined determination angle (NO in step S32), the control ECU 100 sets the target threshold flag to "2" (step S34) and proceeds to step S36. In other words, when the steering angle is smaller than a predetermined judgment angle, the risk of collision with a cyclist is considered to be smaller than when the steering angle is equal to or greater than the predetermined judgment angle, and a side contact monitoring threshold value with lower sensitivity is set than when the steering angle is equal to or greater than the predetermined judgment angle.

[0063] When the control ECU 100 determines that the direction indicator 35 is not ON (determined as NO in step S31), the control ECU 100 sets the side contact monitoring threshold value "1" as the target threshold value flag (step S35), and proceeds to step S36. In other words, when the turn signal 35 is not ON, it is determined that the risk of collision with a cyclist is even lower, and a side contact monitoring threshold value that is less sensitive than when the turn signal 35 is ON is set.

[0064] After setting the side contact monitoring threshold value (step S33, step S34, or step S35), the control ECU 100 executes a pre-vector prediction execution process (step S36). Here, the pre-vector prediction execution process is a process for predicting a vector (predicted vector Y0) indicating a future expected course of the cyclist according to the current riding conditions (past riding conditions: n speed vectors) of the acquired target cyclist without considering the surrounding environment.

[0065] FIG. 6 shows a conceptual diagram of pre-vector prediction. As shown in Fig. 6, in this pre-vector prediction, the average of the target cyclist's speed vector X-1 a certain time ago, the speed vector X-2 an even earlier, and the speed vector X-3 an even earlier than that is taken, and a predicted vector Y0 indicating the predicted cyclist's travel path is calculated. Note that the multiple speed vectors may be weighted in a predetermined manner rather than simply taking the average.

[0066] After executing the pre-vector prediction, the control ECU 100 saves (stores) the predicted vector Y0 of the cyclist (step S37). Specifically, the control ECU 100 stores the calculated predicted vector Y0 in a predicted vector Y0 storage area of ​​the RAM.

[0067] Next, the control ECU 100 calculates a predicted vector Y2 indicating a predicted travel path of the host vehicle from the host vehicle information (step S38). Specifically, the control ECU 100 calculates the predicted vector Y2 of the host vehicle from the host vehicle speed acquired from the VDC 33, the steering angle acquired from the EPS 34, and the like.

[0068] Next, the control ECU 100 predicts a collision prediction time (TTC) from the predicted vector Y0 of the cyclist and the predicted vector Y2 of the vehicle (step S39), and ends the capture-time collision prediction time calculation process.

[0069] 7 shows the concept of the predicted vectors Y0 and Y1 of the cyclist, the predicted vector Y2 of the vehicle, and the predicted time to collision (TTC) prediction concept diagram. Note that the predicted vector Y1 is a corrected version of the predicted vector Y0, and details will be described later. As shown in FIG. 7, the time until collision is predicted based on a predicted vector Y0 or Y1 indicating the predicted path of the cyclist and a predicted vector Y2 indicating the predicted path of the vehicle, and the predicted time to collision (TTC) is calculated.

[0070] (Collision prediction time estimation process) Next, the collision prediction time estimation process performed in step S22 of Fig. 2 will be described with reference to Fig. 4. Fig. 4 is a diagram showing a subroutine of the collision prediction time estimation process.

[0071] In the collision prediction time estimation process, the control ECU 100 first performs a pre-vector prediction execution process (step S51). Here, the pre-vector prediction execution process is the same as the pre-vector prediction execution process in step S36 in the capture collision prediction time calculation process. In other words, the pre-vector prediction execution process predicts a vector (prediction vector Y0) indicating the expected future course of the target cyclist based on the current riding conditions of the cyclist.

[0072] After executing the pre-vector prediction, the control ECU 100 saves (stores) the predicted vector Y0 of the cyclist (step S52). Specifically, the control ECU 100 stores the calculated predicted vector Y0 in a predicted vector Y0 storage area of ​​the RAM.

[0073] Next, the control ECU 100 acquires GPS information (step S53). Specifically, the control ECU 100 acquires vehicle position information from the GPS 31. The control ECU 100 also acquires surrounding geographic information of the vehicle position from the information server 300 using the acquired vehicle position information. That is, by acquiring this GPS information, the vehicle position is specified and used as a source of detection information for surrounding intersections, gradients, and the like. As described above, the surrounding geographic information may be stored in advance in the control ECU 100, and the surrounding geographic information may be acquired from the vehicle position information of the acquired GPS information.

[0074] Next, the control ECU 100 determines whether or not the GPS information has been acquired (step S54). That is, the control ECU 100 determines whether or not the information on the vehicle position, including surrounding geographic information if necessary, has been acquired from the GPS 31.

[0075] If the control ECU 100 determines that the GPS information has been acquired (determined as YES in step S54), it proceeds to step S56, and if it determines that the GPS information has not been acquired (determined as NO in step S54), it proceeds to step S55.

[0076] When it is determined that the GPS information has not been acquired (determined as NO in step S54), the control ECU 100 substitutes the predicted vector Y0 for the predicted vector Y1 (step S55), and proceeds to the process of step S65. Here, the predicted vector Y1 indicates the predicted route of the cyclist, and is the predicted route of the cyclist obtained by correcting the predicted vector Y0 predicted based on the past riding conditions using surrounding geographic information. In other words, when the surrounding geographic information cannot be acquired, the predicted vector Y0 predicted based on the past riding conditions is set as the predicted vector Y1, which is the predicted route of the cyclist after correction.

[0077] On the other hand, when the control ECU 100 determines that the GPS information has been acquired (determined as YES in step S54), the control ECU 100 collates the map data to identify the vehicle position (step S56). That is, the control ECU 100 identifies the position of the vehicle on the map from the vehicle position information (latitude and longitude information) acquired from the GPS 31. Specifically, the control ECU 100 identifies which road and where on the map the vehicle is located. Note that since the captured cyclist is also likely to be traveling parallel to or close to the vehicle, by identifying the vehicle's position, the cyclist can also be determined to be located near the same road.

[0078] Next, the control ECU 100 determines whether or not there is an intersection within a predetermined distance (** [m]) from the current position (step S57). That is, the control ECU 100 refers to the surrounding geographic information and determines whether or not there is an intersection within a predetermined distance from the current vehicle position in the traveling direction.

[0079] If the control ECU 100 determines that an intersection is within the predetermined distance (YES in step S57), it transfers processing to step S58, and if it determines that an intersection is not within the predetermined distance (NO in step S57), it transfers processing to step S59.

[0080] When it is determined that an intersection is present within the predetermined distance (determined as YES in step S57), the control ECU 100 sets a deceleration correction value A (step S58). That is, since there is an intersection ahead, it is determined that the cyclist will slow down or stop, and a predetermined deceleration correction value A is set in a predetermined correction coefficient area. Note that the deceleration correction value A is a correction coefficient used when correcting the cyclist's predicted vector Y0 to the predicted vector Y1. Also, the deceleration correction value A may be a predetermined fixed value, or it may be a value that changes depending on the position of the intersection (the distance from the current position to the intersection).

[0081] Next, when the control ECU 100 sets the deceleration correction value A (step S58) or when it determines that there is no intersection within a predetermined distance (NO in step S57), it determines whether the gradient from the current position to the predicted vector Y0 is equal to or greater than a predetermined gradient rate (+**%) (step S59). That is, the control ECU 100 refers to the surrounding geographic information and determines whether or not the gradient from the current vehicle position to the predicted travel route indicated by the predicted vector Y0 is equal to or greater than a predetermined gradient rate. Note that the gradient information is not limited to the surrounding geographic information, and may be acquired using the gyro sensor 32 or the like. In this way, using the gyro sensor 32 may allow more reliable gradient information to be acquired since real-time values ​​are used.

[0082] If the control ECU 100 determines that the gradient from the current position to the predicted vector Y0 is equal to or greater than a predetermined gradient rate (YES in step S59), it transitions to step S60, and if it determines that the gradient from the current position to the predicted vector Y0 is not equal to or greater than a predetermined gradient rate (NO in step S59), it transitions to step S61.

[0083] When the control ECU 100 determines that the gradient from the current position along the predicted vector Y0 is equal to or greater than a predetermined gradient rate (YES in step S59), the control ECU 100 sets a deceleration correction value B (step S60). That is, since the gradient between the predicted travel paths indicated by the predicted vector Y0 is equal to or greater than a predetermined gradient rate, it is determined that the cyclist will decelerate due to the uphill slope, and a predetermined deceleration correction value B is set. Note that the deceleration correction value B is a correction coefficient used when correcting the predicted vector Y0 of the cyclist to the predicted vector Y1, similar to the deceleration correction value A described above. Also, the deceleration correction value B may be a predetermined fixed value, or may be a value that changes according to the gradient rate.

[0084] Next, when the control ECU 100 sets the deceleration correction value B (step S60) or determines that the gradient from the current position to the predicted vector Y0 is not greater than a predetermined gradient rate (NO in step S59), it determines whether the gradient from the current position to the predicted vector Y0 is less than a predetermined gradient rate (-**%) (step S61). That is, the control ECU 100 refers to the surrounding geographic information and determines whether or not the gradient from the current vehicle position to the predicted travel route indicated by the predicted vector Y0 is equal to or smaller than a predetermined gradient rate. Note that, as with the above, the gradient information may be acquired not only from the surrounding geographic information but also from the gyro sensor 32 or the like.

[0085] If the control ECU 100 determines that the gradient from the current position to the predicted vector Y0 is equal to or less than a predetermined gradient rate (YES in step S61), it transfers processing to step S62, and if the control ECU 100 determines that the gradient from the current position to the predicted vector Y0 is not equal to or less than the predetermined gradient rate (NO in step S61), it transfers processing to step S63.

[0086] When it is determined that the gradient from the current position to the predicted vector Y0 is equal to or smaller than a predetermined gradient rate (YES in step S61), the control ECU 100 sets an acceleration correction value C (step S62). That is, since the gradient between the predicted travel paths indicated by the predicted vector Y0 is equal to or less than a predetermined gradient rate, it is determined that the cyclist will accelerate due to the downhill slope, and a predetermined acceleration correction value C is set. Note that the acceleration correction value C is a correction coefficient used when correcting the cyclist's predicted vector Y0 to the predicted vector Y1, similar to the deceleration correction value A and the deceleration correction value B. Also, the acceleration correction value C may be a predetermined fixed value, similar to the deceleration correction value B, but may also be a value that changes depending on the gradient rate.

[0087] Next, the control ECU 100 sets the acceleration correction value C (step S62) or, if it determines that the gradient from the current position to the predicted vector Y0 is not equal to or less than a predetermined gradient rate (NO in step S61), performs a correction calculation process for the predicted vector Y0 (step S63).

[0088] That is, the control ECU 100 corrects the predicted vector Y0 of the cyclist based on the deceleration correction value A, the deceleration correction value B, and the acceleration correction value C obtained from the surrounding geographic information, and calculates the predicted vector Y1. For example, the control ECU 100 calculates the predicted vector Y1 of the cyclist by the following formula. Y1=Y0*(1+A+B+C) The above A, B, and C indicate acceleration correction values. That is, the above A and B indicate values ​​in which the positive and negative signs of the deceleration correction value A and the deceleration correction value B are reversed, and C indicates the value of the acceleration correction value C.

[0089] Next, the control ECU 100 calculates a predicted vector Y2 indicating a predicted travel path of the host vehicle from the host vehicle information (step S64). Specifically, the control ECU 100 calculates the predicted vector Y2 of the host vehicle from the host vehicle speed acquired from the VDC 33, the steering angle acquired from the EPS 34, and the like.

[0090] Next, the control ECU 100 predicts the collision prediction time (TTC) from the cyclist's predicted vector Y1 and the vehicle's predicted vector Y2 (step S65), and ends the capture-time collision prediction time calculation process. Note that, as described above, when the GPS information cannot be acquired (NO in step S54), the predicted vector Y0 set in step S55 is used as the predicted vector Y1.

[0091] As described above, the collision detection emergency call device 1 equipped with a vehicle surrounding moving object detection device in this embodiment compares the predicted travel path (prediction vector Y2) of the vehicle calculated based on the vehicle's driving information with the predicted travel path (prediction vector Y0 or Y1) of the cyclist calculated based on surrounding object information to calculate the possibility of a collision, changes the judgment threshold for determining whether or not a collision will occur based on the calculated possibility of collision, and detects a collision with the cyclist based on the changed judgment threshold.Therefore, depending on the situation, it is possible to perform collision judgment that prevents erroneous detection even for objects of different types that are to be detected.

[0092] In addition, when a cyclist that had been captured becomes undetectable due to an obstruction, the collision detection emergency notification device of this embodiment corrects the cyclist's predicted trajectory based on surrounding geographic information and calculates the possibility of a collision. Therefore, even if the target moving object is blocked by an obstacle, it is possible to predict the moving object's direction of travel and speed and determine the risk of a collision.

[0093] In this embodiment, the GPS 31, the VDC 33, the EPS 34, and the direction indicator 35 constitute the host vehicle travel information acquisition means of the present application. Also, in this embodiment, the collision detection sensor 20 constitutes the collision detection means of the present application. In this embodiment, the object detection radar 10 constitutes the surrounding object information acquisition means of the present application. In the present embodiment, the control ECU 100 constitutes a travel path prediction control unit, a collision probability calculation control unit, a determination threshold value change control unit, and a surrounding geographic information acquisition control unit of the present invention. [Explanation of symbols]

[0094] 1 collision detection emergency call device, 10 object detection radar, 20 collision detection sensor, 31 GPS, 32 gyro sensor, 33 VDC, 34 EPS, 35 turn indicator, 100 control ECU, 200 emergency call control ECU, 300 information server

Claims

1. A vehicle driving information acquisition means for acquiring vehicle driving information; a collision detection means for detecting a collision between the vehicle and an external object by changing a detection level based on a predetermined judgment threshold; a surrounding object information acquisition means for acquiring information about objects in the vicinity of a traveling position of the vehicle; a travel path prediction control unit that predicts a travel path of the host vehicle based on the driving information of the host vehicle and predicts travel paths of pedestrians or cyclists around the host vehicle based on the object information; a collision possibility calculation control unit that compares a predicted travel path of the host vehicle predicted by the travel path prediction control unit with a predicted travel path of the pedestrian or the cyclist to calculate a collision possibility; a determination threshold change control unit that sets the determination threshold for the collision detection means to determine whether or not a collision exists; a surrounding geographic information acquisition control unit that acquires geographic information including at least one of an intersection and a gradient around the traveling position of the vehicle; Equipped with the determination threshold change control unit sets a determination threshold of a lower value as the collision possibility increases, A vehicle surrounding moving object detection device characterized in that, when the pedestrian or cyclist that was captured by the surrounding object information acquisition means can no longer be captured, the travel path prediction control unit corrects the predicted travel path of the pedestrian or cyclist based on the geographic information acquired by the surrounding geographic information acquisition control unit.

2. The object information used by the travel path prediction control unit to predict a travel path includes: The speed vector information from the time the pedestrian or cyclist was detected to the present is included.

2. The vehicle surrounding moving object detection device according to claim 1.

3. the collision likelihood calculation control unit calculates a collision prediction time based on a predicted travel path of the host vehicle and a predicted travel path of the pedestrian or the cyclist, thereby calculating the collision likelihood; 3. The vehicle surrounding moving object detection device according to claim 1 or 2.

Images