Blind area object detection

Abstract

An object detection system is configured for detecting an object within a front blind area of a vehicle. The object detection system may be configured to determine position coordinates of a plurality of candidate objects in front of a front panel of the vehicle, determine a visibility curve of a driver of the vehicle, and identify each candidate object as one of a visible object or an occluded object based on a comparison of the position coordinates to the visibility curve.

Description

Technical Field

[0001] The present disclosure relates to object detection systems configured to alert an operator of a device of obscured objects, such as, but not necessarily limited to, object detection systems operable to alert a driver of objects found within a vehicle's front blind spot. Background Art

[0002] The driver may be responsible for maneuvering the vehicle to account for objects found within the driver's surroundings. For example, the driver may be tasked with adjusting the vehicle's travel or other maneuvering to avoid an object. While the physical geometry of vehicles may vary widely, some vehicles may be configured so that the driver is positioned primarily facing in a generally forward direction, with various vehicle structures typically positioned further forward of the driver, such that some of these vehicle structures may obscure, block, or otherwise interfere with the driver's field of view. Vehicle structures within the field of view may create blind spots around the vehicle, whereby the driver's ability to observe the corresponding surroundings may be limited to the extent that the driver may not be able to observe objects within the blind spots without adjusting their field of view. Summary of the Invention

[0003] One aspect of the present disclosure relates to an object detection system configured to alert a driver of an obscured object within a vehicle's front blind spot. The object detection system can be configured to classify objects detected near the vehicle as either obscured or visible, so that alerts generated to draw the driver's attention to the object can be limited to obscured objects that the driver may not otherwise be able to see. The ability to limit or avoid alerts for visible objects, when the driver may already be able to see the associated object, can help minimize nuisance and other distractions to the driver.

[0004] One aspect of the present disclosure relates to a method for alerting a driver of an obscured object within a front blind spot of a vehicle. The method may include identifying a plurality of candidate objects within a surrounding environment forward of a front instrument panel of the vehicle, determining position coordinates of each candidate object relative to the vehicle, determining a visibility curve relative to the driver's forward field of view to represent a visibility boundary between the front blind spot and a visible region of the surrounding environment that coincides with the forward field of view, and identifying each candidate object as one of a visible object and an obscured object based on a comparison of the position coordinates with the visibility curve, wherein at least a visible portion of the visible object is within the visible region, and the obscured object has no visible portion within the visible region and / or its entirety is within the front blind spot.

[0005] The method may include, in response to identifying the one or more obscured objects, providing an alert to inform the driver that a hidden object has been discovered within the front blind spot.

[0006] The method may include providing an alert by activating one or more systems on the vehicle to generate at least one of a tactile warning, an audible warning, and / or a visual warning.

[0007] The method may include providing an alert in response to identifying one or more visible objects without providing specific markings or dedicated labels for the visible objects.

[0008] The method may include determining a visibility curve based on a relative comparison of the driver's eye position to the geometry of the vehicle, such that the forward field of view is centered relative to the eye position and the front blind spot corresponds to an area of ​​the forward field of view that is blocked by the geometry.

[0009] The method may include determining a visibility curve based on a relative comparison of the forward visual field to the geometry of one or more of a hood, an instrument panel, an A-pillar, a steering wheel, or other structures of the vehicle in front of the driver, such that a front blind spot corresponds to an area of ​​the forward visual field that is obstructed by one or more of the geometric shapes, and a visible area corresponds to an area of ​​the forward visual field that is not obstructed by the one or more of the geometric shapes.

[0010] The method may include determining position coordinates based on longitudinal distance, lateral distance, and height individually derived for each of the candidate objects from an image of the surrounding environment, the image captured using an imaging device included on the vehicle.

[0011] The method may include the imaging device being a camera having a wide-angle viewing angle and a focal length; determining a mounting position of the camera on a vehicle; and calibrating position coordinates based on the wide-angle viewing angle, the focal length, and the mounting position so as to obtain the position coordinates from images captured using the camera.

[0012] The method may include determining a physical height of each of the candidate objects based on a bounding box derived from the image, and geometrically triangulating the longitudinal distance and the lateral distance using the bounding box.

[0013] The method may include comparing the entity height to a visibility curve; and identifying each of the candidate objects having an entity height above the visibility curve as one of the visible objects.

[0014] The method may include generating the visibility curve such that its slope gradually decreases in a forward direction relative to the front fascia.

[0015] The method may include determining the longitudinal and / or lateral distances based at least in part on ultrasonic distance measurements for the candidate object acquired using an ultrasonic sensor included on the vehicle and image distance measurements for the candidate object derived from an image.

[0016] The method may include defining a relationship between the accuracy of the ultrasound sensor and the imaging device, and determining the longitudinal and / or lateral distance based on selectively weighting the ultrasound and image distance measurements according to the relationship.

[0017] One aspect of the present disclosure relates to a computer-readable storage medium having stored thereon a plurality of non-transitory instructions operable, when executed by one or more processors, to warn a driver of an obscured object within a front blind spot of a vehicle. The non-transitory instructions are operable to identify a plurality of candidate objects within the vehicle's surroundings, determine position coordinates of each candidate object relative to the vehicle, determine a visibility curve for the driver, optionally including a slope that decreases in a forward direction relative to an upper surface of a front fascia of the vehicle, and identify each candidate object as either a visible object or an obscured object based on a comparison of the position coordinates with the visibility curve.

[0018] The non-transitory instructions are operable to determine a physical height of each candidate object and identify a candidate object having a physical height above the slope as one of the visible objects and identify a candidate object having a physical height below the slope as one of the occluded objects.

[0019] The non-transitory instructions are operable to determine a plurality of available visibility curves linked to one or more driver characteristics, determine one or more driver attributes of a driver, and select a visibility curve to correspond to one of the available visibility curves having a driver characteristic that most closely corresponds to the driver attribute and / or interpolate between the available curves.

[0020] The non-transitory instructions are operable to determine a plurality of available visibility curves pegged to one of a plurality of driver eye heights, measure the driver's current eye height using an onboard measurement system, and select a visibility curve to correspond to one of the available visibility curves for which the driver's eye height most closely corresponds to the current eye height and / or interpolate between the available visibility curves.

[0021] One aspect of the present disclosure relates to a vehicle. The vehicle may include a plurality of wheels operable to facilitate movement of the vehicle, a powertrain operable to rotate one or more of the wheels in response to mechanical power generated by an internal combustion engine and / or an electric motor, an imaging system configured to capture images of the surrounding environment in front of a front panel of the vehicle, and an object detection system. The object detection system may be configured to determine position coordinates of a plurality of candidate objects in front of the front panel, determine a visibility curve for a vehicle occupant, and identify each candidate object as one of a visible object or an obscured object based on a comparison of the position coordinates to the visibility curve. The system may also include an alert system configured to provide an alert with a label that is used to draw the occupant's attention to the closest one of the obscured objects.

[0022] The object detection system may be configured to generate a visibility curve to include a shape contour virtually extending in a forward direction relative to an upper surface of the front panel, and to identify candidate objects having a physical height above the shape curve as visible objects, and to identify candidate objects having a physical height below the shape contour as occluded objects.

[0023] The object detection system can be configured to determine the position coordinates based on the longitudinal distance, lateral distance and height obtained individually for each of the candidate objects from the image, and determine the physical height of each of the candidate objects based on geometric triangulation of the longitudinal distance, lateral distance and the height associated therewith.

[0024] These features and advantages and other features and advantages of the present teachings will be readily apparent from the following detailed description of the modes for carrying out the present teachings when taken in conjunction with the accompanying drawings. It will be appreciated that although the following drawings and embodiments may be described separately, their individual features may be combined into additional embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which may be incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0026] Figure 1 An object detection system according to one aspect of the present disclosure is shown.

[0027] Figure 2 A schematic diagram of a vehicle according to one aspect of the present disclosure is shown.

[0028] Figure 3 A flow chart illustrating a method for detecting and alerting a driver to an object according to one aspect of the present disclosure is shown.

[0029] Figure 4Graph illustrating visibility curves according to one aspect of the present invention. DETAILED DESCRIPTION

[0030] As desired, detailed embodiments of the present disclosure may be disclosed herein; however, it is understood that the disclosed embodiments may be merely examples of the present disclosure that may be embodied in various and alternative forms. The drawings may not necessarily be drawn to scale; some features may be exaggerated or minimized to illustrate details of particular components. Therefore, the specific structural and functional details disclosed herein should not be construed as limiting, but merely as a representative basis for teaching those skilled in the art to employ the present disclosure in various ways.

[0031] Figure 1 An object detection system 10 is shown according to one aspect of the present disclosure. The object detection system 10 may be configured to identify and classify objects 11 detected near a vehicle 12 for use in notifying a driver D or other systems on the vehicle 12 of their presence. For exemplary purposes, the vehicle 12 is shown as an automobile, as the present disclosure fully contemplates that the object detection system 10 may be similarly beneficial for other types of vehicles, equipment, etc. Figure 2 As shown in the schematic diagram of FIG, vehicle 12 may include an electric traction motor 14 operable to convert electric power into mechanical power for performing work, such as for mechanically powering a powertrain 16 to propel vehicle 12. Because powertrain 16 optionally includes an internal combustion engine (ICE) 18 for generating mechanical power, vehicle 12 is shown as a hybrid type. Vehicle 12 may alternatively omit the electric motor 14 and / or the ICE 18. Powertrain 16 may include components that facilitate the transmission of mechanical rotational force from traction motor 14 and / or ICE 18 to one or more of wheels 20, 22, 24, and 26. Vehicle 12 may include a rechargeable energy storage system (RESS) 30 to store and supply power to traction motor 14 and / or other components, systems, and the like 32 on vehicle 12, such as via a first bus 34 (e.g., a primary bus or HV bus) and a second bus 36 (e.g., an auxiliary bus or LV bus).

[0032] The vehicle 12 may include a vehicle controller 38 to facilitate monitoring, controlling, measuring, and otherwise directing operations, performance, and the like on the vehicle 12, which may include performing measurements, acquiring readings, or otherwise collecting data to facilitate operations. The vehicle controller 38 may include additional controllers, wherein operations associated therewith are optionally performed in accordance with one or more processors executing corresponding non-transitory instructions stored on one or more computer-readable storage media. An object detection and alert controller 42 may be included to facilitate directing, controlling, or otherwise facilitating the operations and processes described herein for the object detection system 10. The controller 42 may be configured to interact with other systems, controllers, and features included on and / or outside the vehicle to alert the driver, such as by tactile and / or image-based alerts transmitted via an infotainment system, a seat system, or other input / output (I / O) interface 44 adapted to provide corresponding interactions with the driver and / or other occupants, operators, or entities within or outside the vehicle. Although primarily described with respect to alerting the driver D, the present disclosure fully contemplates that the object detection system 10 and / or associated controller 42 may operate in conjunction with an advanced driver assistance system (ADAS), an automated driving system, or other system associated with the vehicle 12. Detected objects 11, object classifications, alerts, etc. may operate in conjunction with such a system to facilitate corrective action without necessarily requiring corresponding communication and / or interaction with the driver.

[0033] While the physical geometry can vary significantly, the illustrated vehicle can be configured such that driver D is positioned primarily facing in a generally forward direction, such that various structures of vehicle 12 can obstruct, block, or otherwise interfere with driver D's field of view 50. Because vehicle 12 is shown moving in a forward direction, driver D's field of view 50 is shown generally corresponding to a forward field of view 50 for illustrative purposes. When considered from the perspective of driver D, the geometry of vehicle 12 structures (such as those associated with one or more of the hood, dashboard, A-pillars, steering wheel, or other structures of vehicle 12, particularly those in front of driver D) can obstruct field of view 50. Vehicle 12 structures within field of view 50 can create associated blind spots around vehicle 12, such that driver D's ability to observe corresponding sectors of the surrounding environment may be limited, i.e., driver D may be unable to observe objects within the blind spots without deviating from their normal driving position. By way of example, the present disclosure will be primarily described with respect to such that a forward blind spot 52 corresponds to an area of ​​forward field of view 50 in front of a front dashboard 54 of vehicle 12. The driver D may be responsible for directing the operation of the vehicle 12 to avoid the object 11 , for example, the driver D may be assigned the task of adjusting the driving direction of the vehicle 12 to avoid the object 11 .

[0034] One aspect of the present disclosure contemplates that object detection system 10 detects object 11 based on information collected using imaging device 58 and / or ultrasonic sensing system 60. Imaging device 58 may be a fisheye camera or other type of recording device with a relatively wide viewing angle and focal length, operable to capture images, video, etc., of the area surrounding vehicle 12. Ultrasonic sensing system 60 or other similar systems may include one or more ultrasonic sensors configured to measure wireless signals reflected from objects 11 near vehicle 12. A controller may utilize the corresponding information collected within imaging device 58, ultrasonic sensor 60, or other detection devices included on vehicle 12 to identify object 11. Object detection system 10 may be configured to classify object 11 as an occluded object 11O or a visible object 11V, depending on whether object 11 is visible or invisible in driver's field of view 50. That is, occluded objects 11O may correspond to objects that are blocked by vehicle 12 structures or otherwise located within one of the blind spots of visible objects 11V, while visible objects 11V may correspond to objects that are visible to driver D. Based on the corresponding classification, the controller 42 can generate an alert to draw the driver D's attention to the object 11, wherein the alert is optionally limited to obscured objects 110 that may not be visible to the driver D. Limiting or avoiding the ability to issue alerts for visible objects 11V (i.e., the ability to optionally limit the alert to obscured objects 11O) can be beneficial in minimizing nuisance and other disruptions to the driver D when the driver D may already have or be assumed to have visibility of the visible object 11V.

[0035] Figure 3 A method flow chart 64 for detecting and alerting a driver D of an object in the vicinity of a vehicle 12 according to one aspect of the present disclosure is shown. The method may be implemented via a detection and alert controller 42 that directs corresponding operations, functions, etc. of the vehicle 12 based on information collected from the object detection system 10 and / or other devices, equipment, sensors, and systems on and / or outside the vehicle 12. The method may be facilitated by one or more processors executing one or more sets of non-transitory instructions stored on corresponding computer-readable storage media. The method is primarily described with respect to an object 11 located within a front blind spot 52 of the vehicle 12, which may generally correspond to an area in front of the vehicle 12 that may not be visible to the driver D due to the geometry of various vehicle structures blocking the driver's field of view 50. As Figure 1 As shown, the front blind spot 52 may correspond to an area of ​​the vehicle 12 that is forward and below an upper surface 66 of the front fascia 54. The front fascia 54 may correspond to a front portion of the vehicle 12 having a bumper, hood, grille, or other structure of the vehicle 12, such as a front fascia 54. Figure 2, which may also include components for an imaging device 58 and / or an ultrasonic sensing system 60. While also beneficial and applicable to other blind spots exterior to the vehicle 12, such as corresponding areas to the rear of the vehicle 12 that may be outside the field of view 50 when the vehicle 12 is in reverse, the method is described with respect to the front blind spot 52 to illustrate the ability of the present disclosure to address the need to decide when to alert the driver D when traveling in a forward direction, particularly because the driver D may generally be more susceptible to and sensitive to alerts when traveling in the forward direction; at least compared to the reverse direction when the vehicle 12 may generally be traveling slower, the driver D may be more likely to desire an alert, or the consequences of drawing the driver's attention may be less significant.

[0036] Block 70 may involve an object detection process, whereby the object detection system 10 may detect a plurality of objects 11 in the vicinity of the vehicle 12, such as within the surrounding environment in front of the front fascia 54. The object detection process may rely on identifying objects 11 detected within images captured by the imaging device 58, signals captured by the ultrasonic sensing system 60, and / or by other mechanisms useful for object detection (e.g., a vehicle perception system). Block 72 may involve a candidate selection process, whereby one or more objects 11 detected during the object detection process may be identified as candidate objects 7. Candidate objects may be a subset of objects 11 detected around the vehicle 12, such as those located in front of the front fascia 54 or within another selected region of interest. The candidate selection process may be used in this manner to filter out other objects detected in other areas of the vehicle 12, for example, to limit the objects considered to those within the path of the vehicle 12 or to eliminate from further processing those objects that may be insignificant with respect to the current operation of the vehicle 12. The selection process may include an identification process 74 whereby each candidate object (which may be referred to for non-limiting purposes as a first object 76, a second object 78, and an Nth object 80) may be assigned an identifier or other discriminator suitable for distinguishing one object from another. For example, a vehicle perception system may be employed to identify and distinguish the candidate objects 76, 78, 80, such as having an object ID (e.g., a number assigned to an object in a frame), an object category (e.g., a person, child, car, truck, boat, etc.), bounding box data, and a probability score.

[0037] The selection process may include coordinate processing 84 for determining positional coordinates for each of the candidate objects 76, 78, and 80, which may correspond to determining a first positional coordinate 86 for the first object 76, a second positional coordinate 88 for the second object 78, and an nth positional coordinate 90 for the nth object 80. The positional coordinates 86, 88, and 90 may be generated relative to the vehicle 12, such as the front instrument panel 54 or other suitable reference (the field of view 50 of the driver D may be related to the other suitable reference). For example, the positional coordinates 86, 88, and 90 may be defined relative to a coordinate system centered on the eyes of the driver D. One aspect of the present disclosure contemplates generating the positional coordinates 86, 88, and 90 based on a longitudinal distance, a lateral distance, and a height obtained for each of the candidate objects 76, 78, and 80 individually from an image captured by the imaging device 58. For example, the longitudinal distance and the lateral distance may be determined based on the mapping capabilities of the imaging device 58, and the height may correspond to the topmost or highest pixel of the associated candidate object relative to a bottom or other reference point in the associated image (e.g., the foot of the object). These values ​​may be geometrically triangulated relative to the position of the imaging camera 58 so that an absolute value, such as the true physical height of the object 11 , may be determined relative to a desired point within the vehicle 12 , such as the driver's eyes.

[0038] In the case where the imaging device 58 is a fisheye, wide-angle, or other type of camera with a predefined or fixed wide-angle viewing angle and focal length, and based on the known mounting location of the fisheye camera on the vehicle 12, the position coordinates 86, 88, 90 of each of the candidate objects 76, 78, 80, 11 can be determined based on the accompanying spatial relationships between the candidate objects 76, 78, 80, 11. The imaging device 58 can undergo a calibration process, whereby the intrinsic properties of the imaging device 58 (such as focal length), non-intrinsic properties (such as mounting height), and the camera lens equation can be used to calculate where each finite area or feature on the ground plane in front of the camera will appear in the image captured by the camera. For each finite area or feature, the longitudinal position of the feature can be recorded in a table at the row / column index corresponding to the pixel coordinates where the finite area appears in the camera image, and the process can be repeated similarly for the lateral position. Alternatively, the longitudinal and / or lateral distances and / or the height of the entity may be based at least in part on ultrasonic distance measurements of the candidate objects 76, 78, 80 using an ultrasonic sensor, such as by defining a relationship between the accuracy of the ultrasonic sensor and the imaging device 58, and determining the distance based on selectively weighting the ultrasonic and imaging distance measurements according to the relationship. For example, the resulting value may be based on combining a weighted value determined from the imaging device 58 with a weighted value determined from the ultrasonic sensing system 60.

[0039] Block 94 involves a closest object detection process, whereby the controller 42 may determine which of the candidate objects 76, 78, and 80 is closest to the front panel 54. This determination may be useful for identifying the object that the vehicle 12 is most likely to approach first if the vehicle 12 continues traveling in the same direction and / or an object is most likely to reach the vehicle 12 first. For example, the objects may be moving, such that the closest object may be temporarily farther away from the vehicle 12 than another object but approaching the vehicle 12 at a higher speed. Block 96 involves a distance evaluation process, whereby the controller 42 may determine whether the closest of the candidate objects 76, 78, and 80 is within a distance threshold from the vehicle 12. The distance threshold may be a parameter that increases and / or decreases depending on the frequency of object detection. For example, the threshold may be increased to reduce noisy activations or hysteresis. If the hysteresis threshold is not exceeded, i.e., if further analysis is desired, block 98 may involve a final determination process, whereby the controller 42 may finalize the candidate object 76, 78, and 80 desired for further blind spot evaluation. Block 100 relates to visibility processing, whereby the controller 42 may categorize each of the candidate objects 76, 78, 80 ultimately determined for evaluation according to visibility to the driver D (e.g., whether the corresponding candidate object is within or outside the front blind spot 52 and / or other blind spots under consideration). Objects 11 determined to be visible may be characterized as visible objects 11V, and objects 11 determined to be occluded (i.e., objects with no visible portion) may be characterized as occluded objects 11O.

[0040] The visibility processing may include identifying each of the candidate objects 76, 78, 80 as one of the visible or obscured objects 110 based on a comparison of the position coordinates 86, 88, 90 associated with each of the candidate objects 76, 78, 80 relative to a visibility curve 102 for the driver D, which may be generated in the visibility curve processing of block 104. The controller 42 may be configured to correspondingly generate a visibility curve 102 for the driver D that may be sufficient to represent the front blind spot 52 and the visible area 106 (see FIG. Figure 1 ) or the area of ​​the surrounding environment within the field of view 50 of the driver D. Figure 4A graph 110 of a visibility curve 102 according to one aspect of the present disclosure is shown. Graph 110 may include a vertical axis 112 representing vertical height and a horizontal axis 114 representing longitudinal distance, which may be defined relative to the front fascia 54 or other reference point associated with the front of the vehicle 12. Because visibility curve 102 may include a lateral shape, e.g., a z-axis may define the lateral shape of visibility curve 102, visibility curve 102 is shown relative to vertical height for simplicity. Visibility curve 102 may be considered a visibility surface, a three-dimensional (3D) shape, etc., which may vary longitudinally and / or laterally relative to field of view 50. For example, the top of hood 66 or other uppermost or frontmost obstruction within field of view 50 may correspond to a peak 120 of visibility curve 102, after which the shape profile of visibility curve 102 may be virtually extended and vary proportionally to the portion or area of ​​the surrounding environment visible to driver D. The visibility curve 102 may generally tilt in a downward manner as it extends beyond the front fascia 54 because the driver D may experience less occlusion when viewing areas further away from the vehicle 12. One aspect of the present disclosure proposes generating multiple available visibility curves to account for different categories of drivers D and / or to determine a personalized visibility curve for each driver D registered or otherwise associated with a particular vehicle. For example, the available visibility curves may be linked to driver D characteristics (such as driver eye height) such that when the vehicle 12 is activated, the corresponding driver D attributes of the current driver D may be determined and used to select the visibility curve 102 from the available visibility curves having the driver D characteristics that most closely correspond to the driver D attributes (i.e., the current eye height of driver D), and / or interpolate between the available visibility curves to match or adjust the available curve for driver D.

[0041] The visibility curve 102 may be generated such that an area 124 above the visibility curve 102 may be considered a visible zone, i.e., a portion of the surrounding environment visible to the driver D, and an area 126 below the visibility curve 102 may be considered an occluded zone, i.e., an area that is occluded or otherwise blocked from view by the driver D. A candidate object 76 , 78 , 80 having a physical height above the visibility curve 102 (such as the first object 128 shown) may be characterized as one of the visible objects 11V, and a candidate object 76 , 78 , 80 having a physical height below the visibility curve 102 (such as the second object 130 shown) may be characterized as one of the occluded objects 11O. A similar analysis can be performed for each additional object in candidate objects 76, 78, 80, so that the results of the visibility processing can include non-visible objects 11V, i.e., classifying each of candidate objects 76, 78, 80 as an occluded object 11O, a mixture of visible objects 11V and occluded objects 11O, or excluding occluded objects 11O (i.e., classifying each of candidate objects 76, 78, 80 as a visible object 11V). The results (i.e., the classification of each of candidate objects 76, 78, 80 as a visible object 11V or an occluded object 11O) can be provided to systems on and / or off-board the vehicle 12 for additional use. The present disclosure proposes a variety of possibilities and enhancements that become available to facilitate the operation of the vehicle based on the advantageous capabilities of the present disclosure, thereby being able to distinguish between visible objects 11V and occluded objects 11O.

[0042] Block 134 relates to full visibility alert processing in response to determining that each of the candidate objects 76, 78, and 80 is a visible object 11V. The full visibility alert processing may include the detection and alert controller 42 providing an alert sufficient to notify the dependent systems that each of the candidate objects 76, 78, and 80 is characterized as a visible object 11V. Because at least a portion of the visible object 11V is within the driver's field of view 50, the alert may result in not warning the driver D or otherwise providing a label to the driver D regarding the visible object 11V. Block 136 relates to full obstruction alert processing in response to determining that each of the candidate objects 76, 78, and 80 is an obstructed object 110. The full obstruction alert processing may include the detection and alert controller 42 providing an alert sufficient to notify the driver D that a hidden object has been detected within the front blind zone 52. The full obstruction alert may include activating one or more systems on the vehicle 12 to generate tactile, audible, and / or visual warnings. The warning may include vibrating the driver D's seat and / or providing a chime, and may include activating a heads-up display or video display to provide a specific marker or dedicated callout to draw the driver's attention to the obscured object 110. Since one of the obscured objects 110 may reach the vehicle 12 before the other obscured objects 110, attention may be focused on that one of the obscured objects 110 as much as possible. Block 138 involves performing a hybrid alert process in response to determining that a portion of the candidate object 76, 78, 80 is an obscured object 110 and a portion of the candidate object 76, 78, 80 is a visible object 11V. The hybrid alert may include activating one or more systems on the vehicle 12 to generate a visual warning, which may include activating a heads-up display or video display to provide a specific marker or dedicated callout to draw the driver's attention to the obscured object 110 or the closest object thereto. The present disclosure contemplates providing a wide variety of alerts, warnings, etc., and thus the foregoing is presented for non-limiting purposes, which may include limiting alerts, warnings, etc. to avoid overwhelming the driver D with alerts, which may include options to limit specific alerts or labeling of certain objects to those objects determined to be occluded objects 110. Block 140 involves the object detection system 10 continuing the foregoing method as the vehicle 12 continues to travel.

[0043] As described above, one aspect of the present disclosure addresses the problem of vehicles often having large blind spots in front of the vehicle due to their shape and height by alerting the driver and automatically presenting a camera view of the area in front of the vehicle if the driver cannot see an object of interest in front of the vehicle. The described detection and alerting methods can be used to generate a real-time estimate of the size of the front blind spot using the vehicle's geometry and the position of the driver's eyes. Using this information, combined with an accurate estimate of the object's position, the system can determine whether the object is visible to the driver and alert the driver accordingly.

[0044] While various embodiments have been described, this description is intended to be illustrative rather than restrictive, and it will be apparent to those skilled in the art that many more embodiments and implementations are possible within the scope of the embodiments. Unless otherwise specified, any feature of any embodiment may be used in combination with or substituted for any other feature or element in any other embodiment. Therefore, the embodiments are not limited except in accordance with the appended claims and their equivalents. Furthermore, various modifications and variations are possible within the scope of the appended claims. Although several modes for implementing many aspects of the present teachings have been described in detail, those skilled in the art to which these teachings relate will recognize various alternative aspects for practicing the present teachings within the scope of the appended claims. It is intended that all content contained in the foregoing description or shown in the accompanying drawings be interpreted as illustrative and exemplary of the entire range of alternative embodiments, which those skilled in the art will recognize as being implied by, structurally and / or functionally equivalent to, or otherwise apparent from the included content, and is not limited solely to those explicitly depicted and / or described embodiments.

Claims

1. A method for warning a driver of an obscured object in a front blind spot of a vehicle, comprising: identifying a plurality of candidate objects within a surrounding environment in front of a front instrument panel of the vehicle; Determine the position coordinates of each candidate object relative to the vehicle; determining a visibility curve relative to a forward visual field of the driver, the visibility curve representing a visibility boundary between the front blind spot and a visible area of ​​the surrounding environment coinciding with the forward visual field; and Based on the comparison of the position coordinates with the visibility curve, each of the candidate objects is identified as one of a visible object or an occluded object, wherein at least a visible portion of the visible object is within the visibility zone, and the occluded object has no visible portion within the visibility zone and / or its entirety is within the front blind zone.

2. The method according to claim 1, further comprising: In response to identifying one or more obscured objects, an alert is provided to inform the driver that a hidden object has been found within the front blind spot.

3. The method according to claim 2, further comprising: The alert is provided by activating one or more systems on the vehicle to generate at least one of a tactile warning, an audible warning, and / or a visual warning.

4. The method according to claim 2, further comprising: In response to identifying one or more visible objects, the alert is provided without providing specific markings or dedicated labels for the visible objects.

5. The method according to claim 1, further comprising: The visibility curve is determined based on a relative comparison of the driver's eye position to a geometry of the vehicle such that the forward field of view is centered relative to the eye position and the front blind spot corresponds to an area of ​​the forward field of view blocked by the geometry.

6. The method according to claim 1, further comprising: The visibility curve is determined based on a relative comparison of the forward visual field with one or more geometric shapes of a hood, an instrument panel, an A-pillar, a steering wheel, or other structures of the vehicle in front of the driver, such that the front blind spot corresponds to an area of ​​the forward visual field that is blocked by one or more of the geometric shapes, and the visible area corresponds to an area of ​​the forward visual field that is not blocked by the one or more of the geometric shapes.

7. The method according to claim 1, further comprising: The position coordinates are determined based on a longitudinal distance, a lateral distance, and a height individually obtained for each of the candidate objects from an image of the surrounding environment, the image being captured using an imaging device included on the vehicle.

8. The method according to claim 7, further comprising: The imaging device is a camera with a wide-angle viewing angle and focal length; determining a mounting location of the camera on the vehicle; as well as The position coordinates are calibrated based on the wide-angle viewing angle, the focal length, and the installation position so as to be obtained from an image captured therewith.

9. The method according to claim 7, further comprising: A physical height of each of the candidate objects is determined based on a bounding box obtained from the image, and the longitudinal distance and the lateral distance are geometrically triangulated using the bounding box.

10. The method according to claim 9, further comprising: comparing the entity height to the visibility curve; as well as Each of the candidate objects having the physical height above the visibility curve is identified as one of the visible objects.

Images