Driver avoidance steering intent detection in vehicle

Abstract

A vehicle system includes a control module configured to receive, from a sensor, data indicative of a torque applied to a steering wheel, calculate a torque rate based on the applied torque, based on the torque, the torque rate and a modeling relationship between the torque and the torque rate, calculating a left evasion steering probability that a driver intends to turn the vehicle to the left and a right evasion steering probability that the driver intends to turn the vehicle to the right; the left evasive steering probability and the right evasive steering probability are compared to determine a directional steering intent of the driver, and a signal is transmitted to a threat evasive steering control module to control the vehicle. Other example vehicle systems and methods for predicting driver intent in evading a steering maneuver are also disclosed.

Description

[0001] introduce

[0002] The information provided in this section is for the purpose of generally presenting the context of the present disclosure. The work of the inventors currently listed (to the extent that it is described in this section) and aspects of the description that may not have been prior art at the time of filing are neither explicitly nor implicitly admitted to be prior art against the present disclosure.

[0003] The present disclosure relates to driver evasive steering intention detection in a vehicle.

[0004] Vehicles often include driver assistance systems. For example, driver assistance systems may include assisted evasive steering, blind spot detection, adaptive cruise control, lane departure warnings, and the like. Assisted evasive steering is a driver-initiated collision avoidance steering mechanism. When assisted evasive steering is employed and the driver begins to steer the vehicle around an object (e.g., another vehicle), the system applies additional steering torque to assist the vehicle in steering around the object. In some cases, the system applies steering torque in the opposite direction after the vehicle passes the object. In such examples, the application of assisted evasive steering is controlled based on vehicle dynamics (e.g., changes in the vehicle's forward movement in response to driver input). Summary of the invention

[0005] A vehicle system for predicting a driver's intention in an evasive steering maneuver, comprising: a sensor configured to monitor a torque associated with a steering wheel in a vehicle; and a control module in communication with the sensor. The control module is configured to: receive data indicating a torque applied to a steering wheel when the vehicle is moving from the sensor, calculate a torque rate based on the applied torque, calculate a left evasive steering probability that the driver intends to steer the vehicle to the left and a right evasive steering probability that the driver intends to steer the vehicle to the right based on the torque, the torque rate, and a modeled relationship between the torque and the torque rate, compare the calculated left evasive steering probability with the calculated right evasive steering probability to determine the driver's directional steering intention, and transmit a signal to a threat avoidance steering control module to control the vehicle. The signal indicates the directional steering intention of the driver.

[0006] In other features, the modeled relationship is a rotated ellipse.

[0007] In other features, the control module is configured to determine a torque offset of the modeled relationship for centering the rotating ellipse about a defined point.

[0008] In other features, the control module is configured to calculate a variance of the torque and set the torque offset to a learned value in response to the variance being less than or equal to a defined threshold.

[0009] In other features, the control module is configured to calculate a variance of the torque and set the torque offset to zero in response to the variance being greater than a defined threshold.

[0010] In other features, the control module is configured to transmit a signal to the threat-avoidance steering control module indicating that the driver's steering intent is to the left in response to the left evasive turn probability being greater than the right evasive turn probability.

[0011] In other features, the control module is configured to transmit a signal to the threat-avoidance steering control module indicating that the driver's steering intent is to the right in response to the right evasive turn probability being greater than the left evasive turn probability.

[0012] In other features, the left evasive turn probability and the right evasive turn probability are based on a weighted average of the torque and the torque rate.

[0013] In other features, the control module is configured to assign a confidence level to the driver's directional turning intent based on one or more defined thresholds.

[0014] In other features, the control module is configured to transmit a signal indicative of a confidence level assigned to the directional steering intent of the driver to the threat-avoidance steering control module.

[0015] In other features, the vehicle includes a vehicle system and a threat avoidance steering control module in communication with the control module.

[0016] A method for predicting a driver's intention in an evasive steering maneuver in a vehicle includes receiving data indicating a torque applied to a steering wheel in a vehicle while the vehicle is moving, calculating a torque rate based on the applied torque, calculating a left evasive steering probability that the driver intends to steer the vehicle left and a right evasive steering probability that the driver intends to steer the vehicle right based on the torque, the torque rate, and a modeled relationship between the torque and the torque rate, comparing the calculated left evasive steering probability to the calculated right evasive steering probability to determine a driver's directional steering intention, and transmitting a signal to a threat avoidance steering control module to control the vehicle. The signal is indicative of the driver's directional steering intention.

[0017] In other features, the modeled relationship is a rotated ellipse, and the method further includes determining a torque offset of the modeled relationship to center the rotated ellipse on a defined point.

[0018] In other features, the method further comprises calculating a variance of the torque and in response to the variance being less than or equal to a defined threshold, setting the torque offset to a learned value.

[0019] In other features, the method further comprises calculating a variance of the torque and setting the torque offset to zero in response to the variance being greater than a defined threshold.

[0020] In other features, transmitting the signal to the threat avoidance steering control module includes transmitting a signal to the threat avoidance steering control module indicating that the driver's steering intent is to the left in response to the left evasive turn probability being greater than the right evasive turn probability.

[0021] In other features, transmitting the signal to the threat avoidance steering control module includes transmitting a signal to the threat avoidance steering control module indicating that the driver's steering intent is to the right in response to the right evasive turn probability being greater than the left evasive turn probability.

[0022] In other features, the left evasive turn probability and the right evasive turn probability are based on a weighted average of the torque and the torque rate.

[0023] In other features, the method further comprises assigning a confidence level to the driver's directional turn intent based on the left evasive turn probability or the right evasive turn probability and one or more defined thresholds.

[0024] In other features, transmitting the signal to the threat avoidance steering control module includes transmitting a signal indicative of a confidence level assigned to the directional steering intent of the driver to the threat avoidance steering control module.

[0025] A vehicle system for predicting a driver's intention in an evasive steering maneuver includes: a sensor configured to monitor a torque associated with a steering wheel in a vehicle; and a control module in communication with the sensor. The control module is configured to: receive data indicating a torque applied to a steering wheel while the vehicle is moving from the sensor, calculate a torque rate based on the applied torque, calculate a left evasive steering probability that the driver intends to steer the vehicle to the left and a right evasive steering probability that the driver intends to steer the vehicle to the right based on the torque, the torque rate, and a modeled relationship between the torque and the torque rate, compare the left evasive steering probability with the right evasive steering probability to determine the driver's directional steering intention, assign a confidence level to the driver's directional steering intention based on the left evasive steering probability or the right evasive steering probability and one or more defined thresholds, and transmit a signal to a threat avoidance steering control module to control the vehicle, the signal indicating the driver's directional steering intention and the corresponding confidence level.

[0026] In other features, the modeled relationship is a rotated ellipse, and the control module is configured to determine a torque offset of the modeled relationship to center the rotated ellipse on a defined point.

[0027] A first aspect of the present disclosure provides a vehicle system for predicting a driver's intention in an evasive steering maneuver, the vehicle system comprising:

[0028] a sensor configured to monitor torque associated with a steering wheel in a vehicle; and

[0029] A control module in communication with the sensor, the control module being configured to:

[0030] receiving data from the sensor indicative of torque applied to a steering wheel while the vehicle is moving;

[0031] calculating a torque rate based on the applied torque;

[0032] calculating a left evasive turn probability that the driver intends to turn the vehicle left and a right evasive turn probability that the driver intends to turn the vehicle right based on the torque, the torque rate, and the modeled relationship between the torque and the torque rate;

[0033] comparing the calculated left evasive turn probability with the calculated right evasive turn probability to determine the driver's directional turn intention; and

[0034] A signal is transmitted to a threat avoidance steering control module to control the vehicle, the signal being indicative of the directional steering intent of the driver.

[0035] According to the vehicle system described in the first aspect of the present disclosure,

[0036] The modeled relationship is a rotated ellipse; and

[0037] The control module is configured to determine a torque offset of the modeled relationship for centering the rotating ellipse at a defined point.

[0038] According to the vehicle system of the first aspect of the present disclosure, the control module is configured to:

[0039] calculating a variance of the torque; and

[0040] In response to the variance being less than or equal to a defined threshold, the torque offset is set to a learned value.

[0041] According to the vehicle system of the first aspect of the present disclosure, the control module is configured as follows:

[0042] calculating a variance of the torque; and

[0043] In response to the variance being greater than a defined threshold, the torque offset is set to zero.

[0044] According to the vehicle system of the first aspect of the present disclosure, the control module is configured as follows:

[0045] In response to the left evasive turn probability being greater than the right evasive turn probability, transmitting a signal to the threat-avoidance steering control module indicating that the driver's steering intention is to the left; and

[0046] In response to the right evasive turn probability being greater than the left evasive turn probability, transmitting a signal to the threat-avoidance steering control module indicating that the driver's steering intention is to the right.

[0047] According to the vehicle system described in the first aspect of the present disclosure, the left evasive steering probability and the right evasive steering probability are based on a weighted average of the torque and the torque rate.

[0048] According to the vehicle system of the first aspect of the present disclosure, the control module is configured to assign a confidence level to the driver's directional turning intention based on one or more defined thresholds.

[0049] According to the vehicle system of the first aspect of the present disclosure, the control module is configured to transmit a signal indicating a confidence level assigned to the directional steering intention of the driver to the threat avoidance steering control module.

[0050] A second aspect of the present disclosure provides a vehicle, comprising the vehicle system according to the first aspect of the present disclosure and a threat avoidance steering control module communicating with the control module.

[0051] A third aspect of the present disclosure provides a method for predicting a driver's intention in an evasive steering maneuver in a vehicle, the method comprising:

[0052] receiving data indicative of torque applied to a steering wheel in the vehicle while the vehicle is moving;

[0053] calculating a torque rate based on the applied torque;

[0054] calculating a left evasive turn probability that the driver intends to turn the vehicle left and a right evasive turn probability that the driver intends to turn the vehicle right based on the torque, the torque rate, and the modeled relationship between the torque and the torque rate;

[0055] comparing the calculated left evasive turn probability with the calculated right evasive turn probability to determine the driver's directional turn intention; and

[0056] A signal is transmitted to a threat avoidance steering control module to control the vehicle, the signal being indicative of the directional steering intent of the driver.

[0057] The method according to the third aspect of the present disclosure, wherein:

[0058] The modeled relationship is a rotated ellipse, and

[0059] The method also includes determining a torque offset of the modeled relationship for centering the rotating ellipse about a defined point.

[0060] The method according to the third aspect of the present disclosure further includes:

[0061] calculating a variance of the torque; and

[0062] In response to the variance being less than or equal to a defined threshold, the torque offset is set to a learned value.

[0063] The method according to the third aspect of the present disclosure further includes:

[0064] Calculate the variance of the torque: and

[0065] In response to the variance being greater than a defined threshold, the torque offset is set to zero.

[0066] According to the method described in the third aspect of the present disclosure, transmitting the signal to the threat avoidance steering control module includes: in response to the left evasive steering probability being greater than the right evasive steering probability, transmitting a signal indicating that the driver's steering intention is to the left to the threat avoidance steering control module.

[0067] According to the method described in the third aspect of the present disclosure, transmitting the signal to the threat avoidance steering control module includes: in response to the right evasive steering probability being greater than the left evasive steering probability, transmitting a signal indicating that the driver's steering intention is to the right to the threat avoidance steering control module.

[0068] The method according to the third aspect of the present disclosure, wherein the left evasive steering probability and the right evasive steering probability are based on a weighted average of the torque and the torque rate.

[0069] The method according to the third aspect of the present disclosure further includes assigning a confidence level to the driver's directional turning intention based on the left evasive turn probability or the right evasive turn probability and one or more defined thresholds.

[0070] The method according to the third aspect of the present disclosure, wherein transmitting the signal to the threat avoidance steering control module comprises transmitting a signal indicative of a confidence level assigned to the directional steering intention of the driver to the threat avoidance steering control module.

[0071] A fourth aspect of the present disclosure provides a vehicle system for predicting a driver's intention in an evasive steering maneuver, the vehicle system comprising:

[0072] a sensor configured to monitor torque associated with a steering wheel in a vehicle; and

[0073] A control module in communication with the sensor, the control module being configured to:

[0074] receiving data from the sensor indicative of torque applied to a steering wheel while the vehicle is moving;

[0075] calculating a torque rate based on the applied torque;

[0076] calculating a left evasive turn probability that the driver intends to turn the vehicle left and a right evasive turn probability that the driver intends to turn the vehicle right based on the torque, the torque rate, and the modeled relationship between the torque and the torque rate;

[0077] comparing the left evasive turn probability with the right evasive turn probability to determine the driver's directional turning intention;

[0078] Assigning a confidence level to the driver's directional steering intention based on the left evasive steering probability or the right evasive steering probability and one or more defined thresholds; and transmitting a signal to a threat avoidance steering control module to control the vehicle, the signal indicating the driver's directional steering intention and the corresponding confidence level.

[0079] According to the vehicle system described in the fourth aspect of the present disclosure,

[0080] The modeled relationship is a rotated ellipse; and

[0081] The control module is configured to determine a torque offset of the modeled relationship for centering the rotating ellipse at a defined point.

[0082] Further areas of applicability of the present disclosure will become apparent from the detailed description, claims and drawings.The detailed description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] The present disclosure will be more fully understood from the detailed description and accompanying drawings, in which:

[0084] Figure 1 is a functional block diagram of an example vehicle system for detecting and predicting driver intent in an evasive steering maneuver in a moving vehicle in accordance with the present disclosure;

[0085] Figure 2 According to the present disclosure, Figure 1 The various parts of the vehicle systems of the vehicle;

[0086] Figure 3 depicts a diagram showing a modeled rotational ellipse in a torque domain having a torque axis and a torque rate axis in accordance with the present disclosure;

[0087] Figure 4 depicts a diagram showing a modeled rotated ellipse in a torque domain having a torque axis and a torque rate axis, wherein the rotated ellipse is translated to be centered about an origin, in accordance with the present disclosure;

[0088] Figure 5 depicts a graph showing bounded regions in a torque domain having a torque axis and a torque rate axis, wherein the bounded regions correspond to different confidence levels according to the present disclosure; and

[0089] Figure 6-7 is a flow chart of an example control process for detecting and predicting driver intent in an evasive steering maneuver according to the present disclosure.

[0090] Among the drawings, reference numerals may be repeated to identify similar and / or identical elements. DETAILED DESCRIPTION

[0091] The vehicle may implement threat avoidance steering (e.g., Assisted Evasion Steering (AES), etc.) to assist the driver in steering around an object, such as another vehicle on the road. In such examples, the threat avoidance steering is a driver-initiated collision avoidance steering mechanism that applies additional steering torque to assist the vehicle in steering around an object. This helps the driver reach the maximum steering angle faster and thus avoid collision with the object. The threat avoidance steering is typically activated based on vehicle dynamics (e.g., changes in the forward movement of the vehicle in response to driver input). However, because this vehicle dynamics-based activation reacts to the movement of the vehicle, assisted steering control with threat avoidance steering may be delayed. This delay in activation, although it may be slight, may cause the vehicle to collide with an object on the road. In addition, in some conditions, such as when the driver is also steering the vehicle around a bend in the road, conventional threat avoidance steering may not be able to adequately detect the movement of the vehicle due to the driver-initiated collision avoidance steering.

[0092] The vehicle system according to the present disclosure is capable of actively and predictively detecting the driver's intention to ensure that the threat avoidance steering is activated and intervened to avoid a collision. For example, the vehicle system monitors the torque applied by the driver and the vehicle state, and effectively detects the driver's evasive maneuver intention in a probabilistic manner. The vehicle system then accurately predicts the driver's evasive steering at the beginning of the maneuver by mathematically modeling the driver's intention in the torque domain. Such technical solutions are capable of reliably detecting and predicting driver-initiated evasive steering, which is effective in all vehicle maneuvers (such as straight and curved maneuvers). Therefore, the vehicle system of the present invention can accurately detect and predict the driver's intention in evasive maneuvers faster (e.g., at least 400 milliseconds faster) and in curves (which are often lost in the case of conventional dynamic-based threat avoidance steering systems) than conventional dynamic-based threat avoidance steering systems.

[0093] Reference now Figure 1 , a block diagram of an example vehicle system 100 for detecting and predicting driver intent in an evasive steering maneuver in a moving vehicle is presented. Figure 1 As shown in FIG, the vehicle system 100 generally includes a control module 102 and a steering sensor 104 in communication with the control module 102. In addition, the vehicle system 100 may optionally include a threat avoidance steering control module 106 in communication with the control module 102 and a detection module 108 in communication with the control module 102 and / or the threat avoidance steering control module 106.

[0094] In various embodiments, the modules and sensors of the vehicle system 100 may share parameters via a network 114, such as a controller (or car) area network (CAN). In such examples, the parameters may be shared via one or more data buses of the network 114. In this way, a given module and / or sensor may make various parameters available to other modules and / or sensors via the network 114.

[0095] Figure 1 The vehicle system 100 may be used in any suitable vehicle, such as an electric vehicle (e.g., a pure electric vehicle, a plug-in hybrid electric vehicle, etc.), an internal combustion engine vehicle, etc. In addition, the system 100 may be applicable to autonomous vehicles, semi-autonomous vehicles, etc. For example, Figure 2 A vehicle 200 is depicted, the vehicle 200 comprising Figure 1 The control module 102 and the threat avoidance steering control module 106, and one or more sensors 204 (e.g., Figure 1 steering sensor 104, etc.).

[0096] Continue to refer Figure 1 , the steering sensor 104 generally monitors the torque associated with the steering wheel in the vehicle. For example, when the driver turns the vehicle to the left or right direction (e.g., evasive steering), the steering sensor 104 can monitor or otherwise sense the torque applied to the steering wheel by the driver. The steering sensor 104 can then transmit the torque (or a representation thereof) to the control module 102 via the network 114. This monitoring and transmission can occur repeatedly, such as periodically, to provide multiple torque values ​​to the control module 102 over time. In such examples, the steering sensor 104 can be any suitable torque sensing device.

[0097] The detection module 108 may generally receive data related to components in the vehicle and / or the environment outside the vehicle. For example, the detection module 108 may collect data indicating detected objects, road conditions, weather conditions, braking, etc. for use in controlling various driver assistance systems (e.g., threat avoidance steering systems, etc.). In various embodiments, the detection module 108 may transmit data indicating detected objects, road conditions, weather conditions, etc. to the threat avoidance steering control module 106 via the network 114 for use in controlling vehicle steering, as further explained herein.

[0098] In various embodiments, the detection module 108 may include one or more detection devices. For example, the detection module 108 may include one or more sensors 110 and / or cameras 112 mounted on the exterior and / or interior of the vehicle. In such examples, the sensors may include any suitable type of sensor, such as a radio radar sensor, a lidar sensor, an ultrasonic sensor, a 3D sensor, etc. The sensor may be a short-range sensor (e.g., a short-range radio radar (SRR) sensor, etc.) and / or a long-range sensor (e.g., a radio radar (LRR) sensor, etc.). In some examples, the camera may include a forward-looking camera, a left-looking camera, and a right-looking camera (relative to the vehicle).

[0099] Figure 1 In an example of the present invention, the control module 102 receives torque data indicating movement of a steering wheel from the steering sensor 104. For example, the control module 102 may receive multiple values ​​(or representations thereof) indicating torque applied to the steering wheel as the vehicle moves over time. In various embodiments, the control module 102 may periodically receive torque-related values. Once received, the control module 102 may combine these values ​​(e.g., the current torque value τ i and one or more previous torque values ​​τ i-1 , τ i- 2. τ i-N ) is temporarily stored in a buffer in the control module 102 or in a buffer associated with the control module 102 for later use.

[0100] The control module 102 may then calculate a torque rate based on the received torque applied to the steering wheel. In various embodiments, the control module 102 may determine the torque rate (eg, the rate at which the torque changes) by calculating a derivative of the torque.

[0101] In various embodiments, the control module 102 may model a relationship for protectively detecting the driver's intent. For example, the control module 102 may model a relationship between one or more torque values ​​and corresponding torque rate values ​​during the initiation of a maneuver (e.g., an evasive maneuver to the right or left). In some examples, the relationship may be modeled as a rotating ellipse in the torque domain.

[0102] For example, Figure 3A diagram 300 is illustrated showing a rotated ellipse 302 in a torque domain having a torque axis 304 (e.g., X-axis) and a torque rate axis 306 (e.g., Y-axis). As shown, the ellipse 302 includes a center 308, a major axis 310, and a minor axis 312. The major axis 310 represents the distance (a) from the center 308 of the ellipse 302 to one vertex of the ellipse 302, and the minor axis 312 represents another shorter distance (b) from the center 308 of the ellipse 302 to another vertex of the ellipse 302. Figure 3 In the example of FIG. 3 , the ellipse 302 is rotated by an angle (θ) 314 .

[0103] The rotating ellipse 302 can be modeled by the following equation (1). As shown in the modeled relationship, τ represents torque, τ' represents torque rate, h is the center of the ellipse on the torque axis 304, k is the center of the ellipse on the torque rate axis 306, θ (theta) is the rotation angle of the ellipse (e.g., angle 314), a represents the distance of the major axis 310 (e.g., the length of one radial axis), and b represents the distance of the minor axis 312 (e.g., the length of another radial axis). In various embodiments, a and b in equation (1) can be values ​​identified (e.g., learned) based on analysis of the modeled relationships for straight-line avoidance maneuvers and curve avoidance maneuvers.

[0104] (1)

[0105] In various embodiments, Figure 1 The control module 102 may detect whether the driver is making an evasive steering maneuver based on, for example, the torque, the torque rate, and the modeled relationship. For example, if the received torque and the calculated torque rate are large enough to fall within a threshold range, the control module 102 may detect an evasive steering maneuver. In such an example, the torque and the torque rate may fall within (according to the modeled relationship) Figure 3 The evasive turn detection window 316 is within. Figure 3 In the example of FIG300, the evasive turn detection window 316 is shown in the first quadrant (Q1) of the diagram 300. This indicates that a left evasive turn is detected. In other examples, another evasive turn detection window may be implemented in the third quadrant (Q3) of the diagram 300, indicating that a right evasive turn is detected.

[0106] In various embodiments, the modeled rotating ellipse may be offset or shifted from the center or origin (0,0) of the torque domain, such as Figure 3. This may occur when the driver initiates an evasive maneuver on a curve in the road, but not when the driver initiates an evasive maneuver on a straight portion of the road. Thus, the ellipse modeled for the straight portion of the road may have a center (h,k) equal to (0,0), while the ellipse modeled for the curved portion of the road may not be centered at (0,0). In such an example, it may be necessary to adjust the rotated ellipse to the center of the torque domain (0,0). This adjustment may be accomplished by introducing an induced torque or an offset torque.

[0107] In some examples, the modeled ellipse of rotation may only be along the torque axis (e.g., Figure 3 In other words, the modeled ellipse can have a center (h, k) equal to (h, 0). In such an example, the induced or offset torque can be based on the center of the ellipse along the torque axis (e.g., equal to h). For example, Figure 4 4 illustrates a diagram 400 showing a rotating ellipse 402, 422 and an evasive turn detection window 416 (eg, similar to FIG. 4 ) in a torque domain having a torque axis 404 (eg, an X-axis) and a torque rate axis 406 (eg, a Y-axis). Figure 3 316). As shown, ellipse 402 (dashed outline) includes a center 408 offset from the center or origin (0,0) of the torque domain. By introducing an offset torque, ellipse 402 can be translated to the center of the torque domain. The translated ellipse is shown as a rotated ellipse 422 (solid outline) with center 410 at the origin (0,0) of the torque domain.

[0108] The control module 102 can determine an offset torque of the modeled relationship that is used to center the rotating ellipse at a defined point (e.g., the center of the torque domain). In doing so, the control module 102 can detect conditions for curved portions of the road (e.g., when the rotating ellipse is not initially centered in the torque domain) using the same general evasive steering detection conditions as employed for straight portions of the road (e.g., when the rotating ellipse is already centered in the torque domain).

[0109] In some examples, equation (1) above may be modified to utilize an offset torque to translate the rotated ellipse to be centered about the origin. For example, the modeled rotated ellipse may be offset only along the torque axis, such as Figure 4 As shown in . In such an example, the following equation (2) represents an example of a modification to equation (1). As shown in equation (2), an offset torque (shown by τinduced) is added to offset the h value. If the offset torque (or induced torque) is equal to h and k is equal to 0 (for example, the modeled ellipse does not deviate from the origin along the torque rate axis), then equation (2) can be simplified to equation (3), as shown below.

[0110] (2)

[0111] (3)

[0112] In various embodiments, the control module 102 determines the offset torque based on various calculations. For example, the control module 102 may calculate the variance (σ) of the applied torque (eg, received from the steering sensor 104). 2 ). The following equation (4) shows the current torque value τ based on a set (N) of torque values ​​( i and one or more previous torque values ​​τ i-1 , τ i-2 , τ i-3 etc.), and more specifically based on the current torque value τ i One way to calculate the variance is to use the average value of the set (N) of torque values. In such an example, the torque values ​​received from the steering sensor 104 may be cached.

[0113] (4)

[0114] Then, the control module 102 can calculate the variance (σ 2 ) to set the offset torque. For example, the control module 102 may set the torque offset to zero in response to the variance being greater than a defined threshold. In other examples, the control module 102 may set the torque offset to a learned value in response to the variance being less than or equal to a defined threshold, setting the torque offset to a learned value (e.g., a value equal to h). In such examples, the learned value may be obtained in any suitable manner, such as using a filter that averages values ​​over time (e.g., a stochastic filter, etc.).

[0115] In addition, the control module 102 calculates the probability of evasive steering to predict the driver's intention. Due to the translation of the ellipse modeled as described above, the calculation can be similar for both the curved part and the straight part of the road. For example, the control module 102 can calculate the probability of evasive steering that the driver intends to turn the vehicle to the left and another probability of evasive steering that the driver intends to turn the vehicle to the right based on the relationship between torque, torque rate and modeling. In such an example, the control module 102 can calculate the left and right evasive steering probability based on the torque and calculate the left and right evasive steering probability based on the torque rate. For example, the following equations (5) and (6) provide a relationship for calculating the right and left evasive steering probability. Specifically, equations (5.1) and (5.2) can be used to calculate the right and left evasive steering probability based on torque (τ), and equations (6.1) and (6.2) can be used to calculate the right and left evasive steering probability based on torque rate (τ').

[0116] (5.1)

[0117] (5.2)

[0118] (6.1)

[0119] (6.2)

[0120] In the examples of equations (5) and (6), β and α represent adjustment parameters associated with torque (τ) or torque rate (τ′), and τ induced More specifically, β represents the sigmoid steepness of the torque (τ) in equation (5) or the torque rate (τ′) in equation (6), and α represents the sigmoid midpoint of the torque (τ) in equation (5) or the torque rate (τ′) in equation (6).

[0121] Then, once the left evasive steering maneuver probability and the right evasive steering maneuver probability for torque and torque rate are calculated, the control module 102 combines the left evasive steering maneuver probability for torque and torque rate and the right evasive steering maneuver probability for torque and torque rate, respectively, as shown in the following equations (7.1) and (7.2). Once combined, the control module 102 can compare the two to determine the direction of the steering intention. For example, if the combined left evasive steering maneuver probability P(AES_left) is greater than the combined right evasive steering maneuver probability P(AES_right), the control module 102 detects that the driver's intention is to evade the left steering. However, if the combined right evasive steering maneuver probability P(AES_right) is greater than the combined left evasive steering maneuver probability P(AES_left), the control module 102 detects that the driver's intention is to evade the right steering.

[0122] (7.1)P(AES_right)=P right (τ)+P right (τ′)

[0123] (7.2) P(AES_left)=P left (τ)+P left (τ′)

[0124] In various embodiments, the evasive steering maneuver probability can be based on a weighted average of torque and torque rate. For example, each evasive steering maneuver probability for torque and torque rate (e.g., calculated by equations (5) and (6) above) can be weighted based on a defined value. In some examples, the evasive steering maneuver probability for torque can be multiplied by a defined weight, and the evasive steering maneuver probability for torque rate can be multiplied by another defined weight. As an example, the following equations (8.1) and (8.2) show the evasive steering maneuver probability P(τ) for torque multiplied by a defined weight (w τ ) and the probability of evasive steering maneuver P(τ′) for the torque rate multiplied by the defined weight (1-w τ ). A weighted evasive steering maneuver probability P(AES_w) may be calculated for both the right probability (Equation (8.1)) and the left probability (Equation (8.2)). The control module 102 may then compare the weighted left evasive steering maneuver probability and the weighted right evasive steering maneuver probability to determine the directional steering intent as explained above.

[0125] (8.1)P(AES_right_w)=w τ P right (τ)+(1-w τ ) right (τ′)

[0126] (8.2)P(AES_left_w)=w τ P left (τ)+(1-w τ ) left (τ′)

[0127] In addition, the control module 102 can determine and assign confidence levels for the driver's directional steering intent based on one or more defined thresholds. In some embodiments, the defined thresholds can correspond to bounded regions. For example, one bounded region can be defined by two thresholds (e.g., a low threshold and a medium threshold), another bounded region can be defined by two thresholds (e.g., a medium threshold and a high threshold), and yet another bounded region can be defined by one threshold (e.g., a high threshold). Figure 5 Illustrated is a graph 500 depicting such a bounded region in a torque domain having a torque axis 504 (eg, X-axis) and a torque rate axis 506 (eg, Y-axis).

[0128] exist Figure 5 , the diagram 500 includes three ovals 510, 512, 514, a rotated ellipse 522 centered at the origin (0, 0), and an evasive turn detection window 516 (e.g., similar to Figure 3 Ovals 510, 512, 514 and evasive turn detection window 516 define bounded regions 530, 540, 550. For example, bounded region 530 is defined by portions of evasive turn detection window 516 and ovals 510, 512, bounded region 540 is defined by portions of evasive turn detection window 516 and ovals 512, 514, and bounded region 550 is defined by evasive turn detection window 516 and oval 514.

[0129] Figure 1The control module 102 can determine whether the modeled ellipse 522 with the torque and the corresponding torque rate value falls within one of the bounded regions 530, 540, 550. If so, the control module 102 can assign a confidence level associated with the driver's directional steering intention. For example, if the torque and the corresponding torque rate value cause the modeled ellipse 522 to fall into the bounded region 530 (e.g., between the ovals 510 and 512), the control module 102 assigns a low confidence level to the driver's directional steering intention. However, if the torque and the corresponding torque rate value cause the modeled ellipse 522 to fall into the bounded region 540 (e.g., between the ovals 512 and 514), the control module 102 assigns a medium confidence level to the driver's directional steering intention. Alternatively, if the torque and the corresponding torque rate value cause the modeled ellipse 522 to fall into the bounded region 550 (e.g., beyond the oval 514), the control module 102 assigns a high confidence level to the driver's directional steering intention.

[0130] The following equations (9), (10), and (11) respectively show the relationship for assigning a low confidence level, a medium confidence level, and a high confidence level to the driver's directional steering intention. In equations (9), (10), and (11), C1 represents a low threshold, C2 represents a medium threshold, and C3 represents a high threshold. In such an example, the threshold C 21 , C2, C3 can correspond to Figure 5 Oval shapes 510, 512, 514.

[0131] (9)

[0132] (10)

[0133] (11)

[0134] Continue to refer Figure 1, the control module 102 may generate one or more signals and transmit them to the threat avoidance steering control module 106. In such an example, the (one or more) signals may indicate different conditions. For example, the control module 102 may provide an indication of the detected driver's intention, an indication of the driver's actual directional steering intention, and / or a confidence level (e.g., low, medium, or high) associated with the directional steering intention. In various embodiments, the control module 102 may additionally and / or alternatively indicate a value associated with a confidence level. For example, if (e.g., weighted) a comparison between the left and right evasive steering maneuver probabilities indicates that the directional steering intention is to the right with a high confidence level, the control module 102 may transmit a signal indicating that the directional steering intention is to the right and a high confidence level. Alternatively, if (e.g., weighted) a comparison between the left and right evasive steering maneuver probabilities indicates that the directional steering intention is to the left with a low confidence level, the control module 102 may transmit a signal indicating that the directional steering intention is to the left and a low confidence level.

[0135] The threat avoidance steering control module 106 may then control the vehicle based on the signal(s), if desired. For example, the threat avoidance steering control module 106 may analyze the received data (e.g., detected driver's intent, the driver's actual directional steering intent, and / or confidence level) and other data received from the detection module 108 and provide a signal 116 for controlling automatic avoidance steering in the vehicle. In some examples, if the confidence level is below a threshold (e.g., a low confidence level or a representative value thereof), the threat avoidance steering control module 106 may ignore the driver's actual directional steering intent or give it a lower weight.

[0136] Figure 6-7 The diagram shows the Figure 1 The vehicle system 100 employs a method for detecting and predicting moving vehicles (e.g., Figure 2 Example control processes 600, 700 of the present invention are provided for controlling the driver's intention in an evasive steering maneuver in a vehicle 200). Although the example control processes 600, 700 are about Figure 1 Although described with respect to the vehicle system 100 including the control module 102 , either of the control processes 600 , 700 may be capable of being employed by any suitable system.

[0137] exist Figure 6 , the control process 600 begins at 602, where the control module 102 receives torque data indicating movement of a steering wheel in a vehicle. In various embodiments, the control module 102 may receive multiple torque values ​​(or representations thereof) indicating torque applied to the steering wheel from a sensor (e.g., the steering sensor 104) over time. Control then proceeds to 604.

[0138] At 604 , the control module 102 calculates or otherwise determines a torque rate associated with the received torque. The control module 102 may determine the torque rate by calculating a derivative of the torque, as explained herein. Control then proceeds to 606 .

[0139] At 606, the control module 102 calculates the left evasive steering maneuver probability and the right evasive steering maneuver probability. In such an example, the control module 102 can determine the evasive steering maneuver probability according to equations (5)-(7) above. In various examples, the evasive steering maneuver probability can be weighted according to equation (8) above. Control then proceeds to 608, 610.

[0140] At 608, the control module 102 compares the left evasive steering maneuver probability and the right evasive steering maneuver probability. At 610, the control module 102 determines which of the evasive steering maneuver probabilities is the greatest. For example, at 610, the control module 102 may determine whether the left evasive steering maneuver probability is greater than the right evasive steering maneuver probability. If "yes" at 610, control proceeds to 612, where the control module 102 generates a signal indicating that the driver's directional steering intent is to the left and transmits it to a threat avoidance steering control module (e.g., Figure 1 However, if “no” at 610 (e.g., the right evasive steering maneuver probability is greater than the left evasive steering maneuver probability), control proceeds to 614, where the control module 102 generates a signal indicating that the driver's directional steering intent is to the right and transmits it to the threat avoidance steering control module. Control may then end, such as Figure 6 as shown in, or return to 602.

[0141] exist Figure 7 In the example of FIG. 7 , control process 700 begins at 702 , where control module 102 receives torque data indicative of movement of a steering wheel in a vehicle, as described above. Control then proceeds to 704 , 716 , and 720 .

[0142] At 704, the control module 102 buffers the multiple values ​​of the received torque. Control then proceeds to 706, 708. The control module 102 calculates a variance based on the buffered torque values ​​at 706 and determines if the calculated variance is less than or equal to a defined threshold at 708. If “yes” at 708, control proceeds to 722. If “no” at 708, control proceeds to 710.

[0143] At 716, the control module 102 calculates or otherwise determines a torque rate associated with the received torque, as explained herein. Control then proceeds to 718. At 718, the control module 102 calculates left and right evasive steering maneuver probabilities based on the calculated torque rates. In various embodiments, the control module 102 may implement equation (6) above to calculate evasive steering maneuver probabilities based on the calculated torque rates. Control then proceeds to 714.

[0144] At 720, the control module 102 learns a torque offset value for translating the modeled rotating ellipse to the origin (0, 0) in the torque domain, as explained herein. For example, the control module 102 may employ a filter (e.g., a random filter, etc.) that averages values ​​over time to learn the torque offset value. Control then proceeds to 722.

[0145] At 722, the control module 102 updates or otherwise sets the torque offset value to the learned value, as explained herein. Thus, if the calculated variance is less than or equal to the defined threshold at 708, the learned value at 720 is used for the torque offset. Control then proceeds to 712.

[0146] At 710, the control module 102 sets (or resets) to zero the torque offset value used to translate the modeled rotating ellipse to the origin (0, 0) in the torque domain. Thus, if the variance calculated at 708 is greater than the defined threshold, the torque offset is set (or reset) to zero. Control then proceeds to 712.

[0147] At 712 , the control module 102 calculates the left and right evasive steering maneuver probabilities based on the torque and torque offset values. In various embodiments, the control module 102 may implement equation (5) above to calculate the evasive steering maneuver probabilities based on the torque and torque offset values. Control then proceeds to 714 .

[0148] At 714, the control module 102 calculates the combined left evasive steering maneuver probability and the combined right evasive steering maneuver probability. For example, the control module 102 can implement equation (7) above to calculate the combined evasive steering maneuver probability. In various examples, the evasive steering maneuver probability can be weighted according to equation (8) above. Control then proceeds to 724, 726.

[0149] At 724, the control module 102 compares the combined left evasive steering maneuver probability and the combined right evasive steering maneuver probability. At 726, the control module 102 determines which of the evasive steering maneuver probabilities is the largest, as described above. If the combined left evasive steering maneuver probability is greater than the combined right evasive steering maneuver probability at 726, control proceeds to 728. Otherwise, if the combined left evasive steering maneuver probability is less than the combined right evasive steering maneuver probability at 726, control proceeds to 732.

[0150] At 728, the control module 102 assigns a confidence level to the combined left evasive steering maneuver probability. In various embodiments, the control module 102 may implement equations (9)-(11) above to assign the confidence level. For example, if the model relationship falls within a first range (e.g., within a low threshold and a medium threshold), the control module 102 may assign a low confidence level. However, if the model relationship falls within a second range (e.g., within a medium threshold and a high threshold), the control module 102 may assign a medium confidence level. If the model relationship falls within a third range (e.g., above a high threshold), the control module 102 may assign a high confidence level. Control then proceeds to 730, where the control module 102 generates a signal indicating that the driver's directional steering intent is to the left and the specified confidence level and transmits it to a threat avoidance steering control module (e.g., Figure 1 The control may then end, such as Figure 7 as shown in, or return to 702.

[0151] At 732, the control module assigns a confidence level to the combined right evasive steering maneuver probability, such as by implementing equations (9)-(11) as described above. Control then proceeds to 734, where the control module 102 generates and transmits a signal indicating the driver's directional steering intent is to the right and the assigned confidence level to the threat avoidance steering control module. Control may then end, such as Figure 7 as shown in, or return to 702.

[0152] The foregoing description is essentially merely illustrative and is in no way intended to limit the present disclosure, its application or use. The broad teachings of the present disclosure can be implemented in a variety of forms. Therefore, although the present disclosure includes specific examples, the true scope of the present disclosure should not be so limited, because other modifications will become apparent after studying the drawings, descriptions and appended claims. It should be understood that one or more steps in the method can be performed in different orders (or simultaneously) without changing the principles of the present disclosure. In addition, although each embodiment is described as having specific features above, any one or more of those features described with respect to any embodiment of the present disclosure can be implemented in any other embodiment and / or combined with the features of any other embodiment, even if the combination is not explicitly described. In other words, the described embodiments are not mutually exclusive, and the arrangement of one or more embodiments to each other is still within the scope of the present disclosure.

[0153] The spatial and functional relationships between elements (e.g., between modules, circuit elements, semiconductor layers, etc.) are described using various terms, including "connected," "engaged," "coupled," "adjacent," "next to," "on top of," "above," "below," and "disposed." Unless explicitly described as "directly," when describing the relationship between a first element and a second element in the above disclosure, the relationship can be a direct relationship, in which there are no other intermediate elements between the first element and the second element, but can also be an indirect relationship, in which there are one or more intermediate elements (spatially or functionally) between the first and second elements. As used herein, the phrase "at least one of A, B, and C" should be interpreted to mean a logical (A or B or C), using a non-exclusive logical "or," and should not be interpreted to mean "at least one of A, at least one of B, and at least one of C."

[0154] In the drawings, the direction of the arrows, as indicated by the arrows, generally demonstrates the flow of information (e.g., data or instructions) of interest to the illustration. For example, when component A and component B exchange various information, but the information transmitted from component A to component B is relevant to the illustration, the arrow may be directed from component A to component B. This unidirectional arrow does not mean that no other information is transmitted from component B to component A. In addition, for information sent from component A to component B, component B may send a request or a receipt confirmation of the information to component A.

[0155] In this application, including the definitions below, the term "module" or the term "controller" may be replaced with the term "circuit". The term "module" may refer to, be part of, or include: an application specific integrated circuit (ASIC); a digital, analog, or mixed analog / digital discrete circuit; a digital, analog, or mixed analog / digital integrated circuit; a combinational logic circuit; a field programmable gate array (FPGA); a processor circuit (shared, dedicated, or group) that executes code; a memory circuit (shared, dedicated, or group) that stores code executed by the processor circuit; other suitable hardware components that provide the described functionality; or a combination of some or all of the above, such as in a system on a chip.

[0156] The module may include one or more interface circuits. In some examples, the interface circuit may include a wired or wireless interface connected to a local area network (LAN), the Internet, a wide area network (WAN), or a combination thereof. The functions of any given module of the present disclosure may be distributed between multiple modules connected via the interface circuit. For example, multiple modules may allow load balancing. In a further example, a server (also referred to as a remote or cloud) module may perform some functions on behalf of a client module.

[0157] The term "code" as used above may include software, firmware, and / or microcode, and may refer to a program, a routine, a function, a class, a data structure, and / or an object. The term "shared processor circuit" encompasses a single processor circuit that executes some or all of the code from multiple modules. The term "group processor circuit" encompasses a processor circuit that executes some or all of the code from one or more modules in combination with an additional processor circuit. References to multiple processor circuits encompass multiple processor circuits on discrete dies, multiple processor circuits on a single die, multiple cores of a single processor circuit, multiple threads of a single processor circuit, or a combination of the above. The term "shared memory circuit" encompasses a single memory circuit that stores some or all of the code from multiple modules. The term "group memory circuit" encompasses a memory circuit that stores some or all of the code from one or more modules in combination with additional memory.

[0158] The term "memory circuit" is a subset of the term "computer-readable medium". As used herein, the term "computer-readable medium" does not cover transient electrical or electromagnetic signals propagated through a medium (such as on a carrier wave); therefore, the term "computer-readable medium" may be considered to be tangible and non-transitory. Non-limiting examples of non-transitory, tangible computer-readable media are non-volatile memory circuits (such as flash memory circuits, erasable programmable read-only memory circuits, or mask read-only memory circuits), volatile memory circuits (such as static random access memory circuits or dynamic random access memory circuits), magnetic storage media (such as analog or digital magnetic tape or hard disk drives), and optical storage media (such as CDs, DVDs, or Blu-ray discs).

[0159] The apparatus and methods described in this application may be implemented in part or in whole by a special purpose computer created by configuring a general purpose computer to perform one or more specific functions embodied in a computer program. The above-mentioned function blocks, flow chart components and other elements serve as software specifications, which can be translated into a computer program by routine work of a skilled technician or programmer.

[0160] The computer program includes processor executable instructions stored on at least one non-transitory tangible computer readable medium. The computer program may also include or rely on stored data. The computer program may include a basic input / output system (BIOS) that interacts with the hardware of the special-purpose computer, device drivers that interact with specific devices of the special-purpose computer, one or more operating systems, user applications, background services, background applications, etc.

[0161] A computer program may include: (i) descriptive text to be parsed, such as HTML (Hypertext Markup Language), XML (Extensible Markup Language), or JSON (JavaScript Object Notation), (ii) assembly code, (iii) object code generated by a compiler from source code, (iv) source code for execution by an interpreter, (v) source code for compilation and execution by a just-in-time compiler, etc. By way of example only, the source code may be written using syntax from a language including: C, C++, C#, Objective C, Swift, Haskell, Go, SQL, R, Lisp, Fortran, Perl, Pascal, Curl, OCaml, HTML5 (Hypertext Markup Language Fifth Edition), Ada, ASP (Active Server Pages), PHP (PHP: Hypertext Preprocessor), Scala, Eiffel, Smalltalk, Erlang, Ruby, Visual Lua, MATLAB, SIMULINK, and

Claims

1. A vehicle system for predicting a driver's intention in an evasive steering maneuver, the vehicle system comprising: a sensor configured to monitor torque associated with a steering wheel in a vehicle; as well as A control module in communication with the sensor, the control module being configured to: receiving data from the sensor indicative of torque applied to a steering wheel while the vehicle is moving; calculating a torque rate based on the applied torque; calculating a left evasive turn probability that the driver intends to turn the vehicle left and a right evasive turn probability that the driver intends to turn the vehicle right based on the torque, the torque rate, and the modeled relationship between the torque and the torque rate; comparing the calculated left evasive turn probability with the calculated right evasive turn probability to determine the driver's directional turning intention; as well as A signal is transmitted to a threat avoidance steering control module to control the vehicle, the signal being indicative of the directional steering intent of the driver.

2. The vehicle system according to claim 1, wherein The modeled relationship is a rotated ellipse; and The control module is configured to determine a torque offset of the modeled relationship for centering the rotating ellipse at a defined point.

3. The vehicle system of claim 2, wherein the control module is configured to: calculating a variance of the torque; and In response to the variance being less than or equal to a defined threshold, the torque offset is set to a learned value.

4. The vehicle system of claim 2, wherein the control module is configured to: calculating a variance of the torque; and In response to the variance being greater than a defined threshold, the torque offset is set to zero.

5. The vehicle system of claim 1 , wherein the control module is configured to: In response to the left evasive turn probability being greater than the right evasive turn probability, transmitting a signal to the threat-avoidance steering control module indicating that the driver's steering intention is to the left; and In response to the right evasive turn probability being greater than the left evasive turn probability, transmitting a signal to the threat-avoidance steering control module indicating that the driver's steering intention is to the right. 6 . The vehicle system of claim 5 , wherein the left evasive turn probability and the right evasive turn probability are based on a weighted average of the torque and the torque rate. 7 . The vehicle system of claim 6 , wherein the control module is configured to assign a confidence level to the driver's directional steering intent based on one or more defined thresholds. 8 . The vehicle system of claim 7 , wherein the control module is configured to transmit a signal indicative of a confidence level assigned to the driver's directional steering intent to the threat avoidance steering control module.

9. A vehicle comprising the vehicle system of claim 1 and a threat avoidance steering control module in communication with the control module.

10. A method for predicting a driver's intention in an evasive steering maneuver in a vehicle, the method comprising: receiving data indicative of torque applied to a steering wheel in the vehicle while the vehicle is moving; calculating a torque rate based on the applied torque; calculating a left evasive turn probability that the driver intends to turn the vehicle left and a right evasive turn probability that the driver intends to turn the vehicle right based on the torque, the torque rate, and the modeled relationship between the torque and the torque rate; comparing the calculated left evasive turn probability with the calculated right evasive turn probability to determine the driver's directional turning intention; as well as A signal is transmitted to a threat avoidance steering control module to control the vehicle, the signal being indicative of the directional steering intent of the driver.

Images