Mobile carrier lateral blind spot sensing system and method thereof

Abstract

The present application provides a kind of mobile vehicle lateral blind spot sensing system and its method, which is applied to a vehicle when traveling to a parking position, first using a light scanning unit and an image capturing unit to scan the objects corresponding to the parking position and capture a plurality of object images corresponding to the objects, to filter out a plurality of corresponding filtered images by the travel route of the vehicle, and further obtain a corresponding predicted trajectory, thereby generating a route modification to inform the driving or assist driving system, since the objects mapped by the filtered images are located at a blind spot position on one side of the vehicle, so that the driver avoids ignoring the lateral blind spot and ignores the danger.

Description

Technical Field

[0001] This invention relates to a blind spot sensing system and method, and more particularly to a lateral blind spot sensing system and method for a mobile vehicle. Background Technology

[0002] Traditional Advanced Driver Assistance Systems (ADAS) are developed to assist drivers and can be broadly divided into three main parts: onboard sensors, onboard processors, and actuators. ADAS uses onboard sensors to detect signals outside the vehicle. These sensors include millimeter-wave radar and LiDAR, as well as thermal and pressure sensors. The sensor data is transmitted back to the onboard processor, such as the Electronic Control Unit (ECU), which generates warning messages that the driver can recognize to avoid dangerous road conditions. In cases where the driver's reaction is insufficient, the onboard sensors may even directly intervene in the driver's driving behavior and activate actuators to achieve functions such as vehicle deceleration, emergency braking, or vehicle swerving to protect the driver.

[0003] Furthermore, to protect drivers, manufacturers have developed radar detection technology to detect the area around vehicles. However, radar cannot identify whether the area around a vehicle consists of fixed obstacles or moving objects. Moreover, when objects that do not affect the vehicle's movement approach the vehicle, the warning unit will still frequently issue warning messages, which also causes inconvenience to drivers. Although there have been many improvements to the detection of obstacles around moving vehicles to achieve monitoring, in addition to the dangers posed by other vehicles, there are many more people and things that can affect the movement of a moving vehicle. For example, pedestrians, animals, and moving objects can be considered obstacles to moving vehicles, which can cause emergency situations during the journey. This impact is most serious in congested urban streets.

[0004] While manufacturers have developed color image capture technologies such as dashcams to record the situation in emergency situations, this is not a solution in hindsight. The real solution lies in helping drivers prevent emergencies from occurring in the first place. Currently, dashcams are only installed on the front and rear sides of the vehicle, leaving blind spots. Therefore, it is necessary to further assist drivers in avoiding blind spots by integrating side imaging equipment and detection technology. Furthermore, it is necessary to anticipate dangers and notify drivers based on the detection of side blind spots in order to protect them.

[0005] However, dangerous situations can occur not only at intersections, but also when a vehicle is entering a parking space. Especially with the widespread use of automatic parking technology today, it is no longer enough for ADAS to intervene in the driver's driving behavior to protect the driver; it is also necessary to anticipate dangers in advance.

[0006] To address the aforementioned problems, this invention provides a lateral blind spot sensing system and method for mobile vehicles. The system scans objects on one side of the vehicle to obtain corresponding object images, filters out images that affect the vehicle, and generates a predicted path based on the objects mapped from the filtered images, thereby modifying the corresponding travel route to avoid dangerous situations. Summary of the Invention

[0007] One objective of this invention is to provide a lateral blind spot sensing system and method for a mobile vehicle, which obtains multiple object images by scanning objects on one side of the vehicle, and then filters out the images corresponding to the lateral blind spot of the vehicle to obtain the predicted movement line of the corresponding object, thereby generating a corresponding modification of the travel route to avoid dangerous situations.

[0008] To achieve the above objectives, the present invention discloses a method for lateral blind spot sensing of a mobile vehicle. This method is applied to a vehicle equipped with a host computer connected to an optical scanning unit and an image capturing unit. The host computer executes the following steps: first, it executes a parking command corresponding to the vehicle, causing the vehicle to move into a corresponding parking position; then, it generates a positioning message based on the relative or absolute position of the vehicle and the parking position; next, it generates a first travel route based on the positioning message and the parking position; and finally, based on the first travel route, it uses the optical scanning unit to scan the area around the parking position. The system identifies at least one object and captures an image of that object using the image capturing unit. Next, the system classifies the object image using an image optical flow method to obtain at least one filtered image corresponding to the parking position. Then, the system generates at least one predicted movement path based on the object vector mapped from the at least one filtered image, i.e., predicts the movement path of the objects mapped from the filtered images. Following this, the system modifies the first travel route based on the at least one predicted movement path and generates a corresponding second travel route. In other words, the system performs hazard prediction for the vehicle's blind spots and adjusts the vehicle's corresponding travel route. Therefore, this invention provides hazard prediction for the vehicle's lateral blind spots during the parking phase and generates a corresponding modified travel route, thereby enabling the driver assistance system to intervene in driving control based on notification messages and simultaneously notify the driver.

[0009] The present invention provides an embodiment in which, in the steps of scanning at least one object corresponding to the parking position using the optical scanning unit according to the first travel route and capturing an image of the corresponding at least one object using the image capturing unit, the optical scanning unit further scans the at least one object around the parking position and captures an image of the corresponding at least one object around the parking position using the image capturing unit.

[0010] The present invention provides an embodiment in which, in the step of classifying the object images according to the first travel route and the parking position using an image optical flow method, the host extracts multiple three-dimensional images based on the filtered images and classifies the object images according to the positioning information combined with the image optical flow method.

[0011] The present invention provides an embodiment in which, in the step of modifying the first travel route according to the at least one predicted movement line and correspondingly generating a second travel route, the host determines whether a first effective area of ​​the parking position is reduced to a second effective area according to the at least one predicted movement line. The first effective area is larger than the vehicle size, and the second effective area is smaller than the vehicle size. When the first effective area is reduced to the second effective area, the second travel route indicates that the vehicle is parked in part of the parking position.

[0012] The present invention provides an embodiment in which, in the step of modifying the first travel route and generating a corresponding second travel route based on the at least one predicted travel line, the host calculates based on an inner wheel difference and a steering angle corresponding to the first travel route and the at least one predicted travel line, thereby modifying the first travel route and generating the corresponding second travel route.

[0013] The present invention further provides a lateral blind spot sensing system for a mobile vehicle, wherein the mobile vehicle lateral blind spot sensing system includes a host, a positioning unit, a light scanning unit and an image capturing unit, wherein the host is disposed inside the vehicle, and the light scanning unit and the image capturing unit are disposed on one side of the vehicle. The host executes a parking command based on a parking position corresponding to the vehicle, and generates a positioning message based on the relative or absolute position of the vehicle and the parking position. The host then generates a first travel route based on the positioning message and the parking position. Based on the first travel route, the host uses the optical scanning unit to scan multiple objects corresponding to that side of the vehicle, and the image capturing unit captures the image of at least one object corresponding to the scanned object and sends it to the host. Therefore, the host uses an image optical flow method to filter the object images to obtain at least one filtered image, and the host generates at least one predicted movement line based on the at least one filtered image. In this way, the host modifies the first travel route based on the at least one predicted movement line to generate a second travel route. Since the objects mapped by the filtered images are located in a blind spot on one side of the vehicle, the host generates a corresponding travel route modification by predicting the danger situation in the blind spot, which can be used to notify the driver assistance system to intervene or to notify the driver.

[0014] The present invention provides an embodiment in which the optical scanning unit is a light source or a laser scanner.

[0015] The present invention provides an embodiment in which the optical scanning unit further scans the at least one object around the parking position and uses the image capturing unit to capture an image of the at least one object corresponding to the parking position.

[0016] The present invention provides an embodiment in which the host determines, based on the at least one predicted movement path, whether a first effective area of ​​the parking position is reduced to a second effective area, the first effective area being larger than a vehicle size of the vehicle, and the second effective area being smaller than the vehicle size. When the first effective area is reduced to the second effective area, the second movement path indicates that the vehicle is parked in part of the parking position.

[0017] The present invention provides an embodiment in which the host calculates based on an inner wheel difference and a steering angle corresponding to the first travel route and the at least one predicted movement line, thereby modifying the first travel route and generating the second travel route accordingly.

[0018] The present invention provides an embodiment in which the lateral blind spot location corresponds to a blind spot area of ​​the vehicle relative to the parking position and conforms to the ISO 17387 standard for intelligent transportation system certification. Attached Figure Description

[0019] Figure 1 This is a flowchart of an embodiment of the present invention;

[0020] Figures 2A to 2F This is a schematic diagram of some steps in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of a perspective projection method according to an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of the parking position of the parking section according to an embodiment of the present invention; and

[0023] Figure 5 This is a schematic diagram of a vehicle parked in a parking space according to an embodiment of the present invention.

[0024] [Figure Number Reference Guide]

[0025] 1 Detection System

[0026] 10 hosts

[0027] 12 processing units

[0028] 14MB RAM

[0029] 20 optical scanning units

[0030] 30 image capturing units

[0031] 40 positioning units

[0032] 42 Location Messages

[0033] 50 parking spaces

[0034] CMD Stop Command

[0035] DM1 First Region

[0036] DM2 second region

[0037] IMG Image Screening

[0038] L-image optical flow method

[0039] L1 First Route

[0040] L1D First Move Data

[0041] L2 Second Route

[0042] L2D second row data

[0043] ML Prediction Flow

[0044] MLD Motion Data

[0045] OBJ object images

[0046] P operation procedure

[0047] P0 image point

[0048] P1 First Image Point

[0049] P2 Second Image Point

[0050] V-vehicle

[0051] V3D 3D Images

[0052] VO1 First Object

[0053] VO2 Second Item

[0054] x1 First X-axis

[0055] x2 Second X-axis

[0056] y1 First X-axis

[0057] y2 second X-axis

[0058] Steps S10 to S20 Detailed Implementation

[0059] To provide a better understanding of the structural features and effects achieved by the present invention, preferred embodiments and detailed descriptions are provided below:

[0060] In view of the fact that conventional radar systems and dashcams fail to provide lateral blind spot prediction for vehicles, the present invention proposes a lateral blind spot sensing system and method for mobile vehicles to solve the problem that conventional technology makes it difficult to avoid dangerous situations involving lateral blind spots of vehicles.

[0061] The following will further explain the characteristics provided by the lateral blind spot sensing system and method for mobile vehicles disclosed in this invention, as well as the associated system:

[0062] First, please refer to Figure 1 This is a flowchart of an embodiment of the present invention. As shown in the figure, the method for lateral blind spot sensing of a mobile vehicle according to the present invention shall be performed on a host computer in the following steps:

[0063] Step S10: Determine whether the vehicle has turned and moved to a parking position;

[0064] Step S12: Generate positioning information based on the relative or absolute position of the vehicle and the parking location;

[0065] Step S122: The host generates a first travel route based on the positioning information and the location information of the parking position;

[0066] Step S14: Based on the first travel route, use the light scanning unit to scan the corresponding objects at or near the parking position and use the image capturing unit to capture the corresponding object images;

[0067] Step S16: Classify object images using image optical flow method to obtain the filtered image corresponding to the first travel route;

[0068] Step S18: Generate predicted motion lines based on the object vectors mapped from the filtered image; and

[0069] Step S20: Adjust the first travel route according to the predicted movement line and generate the corresponding second travel route.

[0070] Please refer to the following: Figures 2A to 2EThe identification system 1 used in conjunction with the method for lateral blind spot sensing of mobile vehicles according to the present invention includes a host 10, a light scanning unit 20, and an image capturing unit 30. In this embodiment, the host 10 is exemplified by an automotive computer having a processing unit 12 and a memory 14, but it is not limited to this. It can also be a server, a laptop computer, a tablet computer, or an electronic device with basic image processing capabilities, all of which are host 10 as referred to in the present invention. In this embodiment, the light scanning unit 20 is a LiDAR device or a laser scanner. In this embodiment, the image capturing unit 30 is a color image capturing unit, such as an automotive CMOS image sensor. The host 10 executes a calculation program P through the processing unit 12, which is used to receive the image data IMG generated by the image capturing unit 30 and perform image processing. The host 10 is disposed within a carrier V. The optical scanning unit 20 and the image capturing unit 30 are disposed on one side of the carrier V. The host 10 is electrically connected to the optical scanning unit 20 and the image capturing unit 30. In this embodiment, the image capturing unit 30 has an image capturing angle range of 120 to 170 degrees and captures images within a 10-meter radius of the carrier V, such as object images, as detailed later. Furthermore, the host 10 is electrically connected to a positioning unit 40.

[0071] In step S10, as Figure 2A As shown, host 10 determines whether a stop command CMD has been executed. That is, host 10 determines whether vehicle V turns and moves towards a parking position 50. If no stop command is executed, it continues to determine whether a stop command is executed, that is, it re-executes step S10. If a stop command CMD has been executed, it continues to execute step S12. See also Figure 2A and Figure 2B In this embodiment, a positioning message 42 generated by the positioning unit 40 is transmitted to the processing unit 12 of the host 10. The positioning unit 40 generates the positioning message 42 to the processing unit 12 based on the absolute position of the vehicle V and the parking position 50. The processing unit 12 then proceeds to step S122 to generate a first travel route L1 corresponding to the vehicle V based on the positioning message 42 and the parking position 50. This first travel route L1 is an example of the vehicle V turning to the parking position 50. This first travel route L1 is a default route for the vehicle V to travel to the parking position 50. Therefore, in this embodiment, step S14 is executed next. In addition to using the positioning unit 40 to provide absolute position positioning information 42, the present invention can also use the optical scanning unit 20 to perform optical scanning on one side of the vehicle V or even scan within ten to fifty meters around the vehicle V to provide relative position positioning information 42 of the parking position 50. That is, the optical scanning unit 20 obtains relative positioning results of the surrounding space of the vehicle V, thereby providing positioning information 42 corresponding to the parking position 50 and relative to the vehicle V.

[0072] Host 10 executes step S14, see reference. Figure 2A and Figure 2B The host 10 uses the optical scanning unit 20 to optically scan one side of the vehicle V according to the first travel route L1, especially the parking position 50, and can further scan the front and rear surroundings of the parking position 50. That is, the optical scanning unit 20 scans the objects corresponding to the parking position 50. The scanning method of the optical scanning unit 20 is to generate at least one grating 22 to at least one object. In this embodiment, the objects are the first object VO1 and the second object VO2 as examples. The first object VO1 and the second object VO2 will be scanned according to the grating. The optical scanning unit 22 generates reflected light 32, which is reflected to the image capturing unit 30, thereby generating multiple object images OBJ. In this embodiment, the optical scanning unit 20 is exemplified by a light source, which uses a plurality of parallel light rays as a grating 22, especially vertically parallel light rays, such as parallel laser light. Thus, the image capturing unit 30 captures the reflected light 32 corresponding to the grating 22 to generate the object image OBJ corresponding to the reflected light 32. In addition, the optical scanning unit 20 of the present invention can also be a laser scanner, achieving the effect of light source by multiple laser light scans. The processing unit 12 performs preprocessing on the object image OBJ captured by the image capturing unit 30 by executing the calculation program P, thereby highlighting the corresponding first object VO1 and second object VO2 on the object image OBJ, and performing image stitching and color grayscale correction on the object image OBJ to provide subsequent spatial recognition.

[0073] The lateral blind spot is a blind spot area corresponding to the vehicle relative to the parking position and conforming to the ISO 17387 standard for intelligent transportation system certification. The light scanning unit 20 specifically targets the visual blind spot position on the parking position 50 relative to the vehicle V where the first object VO1 or the second object VO2 is not visually visible, that is, the blind spot position outside the driver's forward visual area. Even if the vehicle V has left and right rearview mirrors, the auxiliary light scanning unit 20 and image capturing unit 30 are still needed to capture the undetected areas. The automatic driving assistance system also needs more complete image capturing to more accurately identify whether there are objects on the side of the vehicle V, such as people, vehicles, bus stop signs, traffic signs, or traffic signals, or even the A-pillar inside the vehicle, which is a visual position where visual blind spots often occur.

[0074] Following step S16, as follows Figure 2CAs shown, processing unit 12 performs an image optical flow method L to filter object images OBJ and selects a filter image IMG. That is, it filters the corresponding objects based on the first travel path L1 of vehicle V, thus obtaining the corresponding filter image IMG. For example, if an object is parked on the roadside or is a vehicle, processing unit 12 does not consider its corresponding object image OBJ and does not mark it as one of the filter images. Figure 2B As shown, object VO includes a first object VO1 and a second object VO2. The second object VO2 is a vehicle parked on the roadside, and therefore does not affect the first travel route L1 of vehicle V. Thus, the object image OBJ of the second object VO2 is not marked as a filter image IMG, meaning the object image OBJ of the first object VO1 is filtered as a filter image IMG. In this embodiment, the processing unit 12 extracts the three-dimensional image V3D of the first object VO1 by executing the calculation program P, and performs spatial identification based on the three-dimensional image V3D. That is, the host 10 uses the positioning information 42 provided by the positioning unit 40 and performs spatial identification based on the three-dimensional image V3D, thus confirming that the second object VO2 is a parked vehicle that has not moved. Furthermore, the first object VO1 in this embodiment can be a road user riding in vehicle V, but it is not limited to this; it can also be a moving vehicle.

[0075] See also step S18. Figure 2B and Figure 2D The host 10 executes the calculation program P to perform a prediction calculation based on the filtered image IMG, and predicts the predicted movement line ML of the first object VO1 corresponding to the filtered image IMG. The processing unit 12 performs the prediction calculation based on the positioning information 42 and the object vector corresponding to the filtered image IMG, thus calculating the movement line data MLD corresponding to the filtered image IMG, which corresponds to... Figure 2B The predicted path ML shown can be represented as follows: the object vector corresponding to the filtered image IMG can be 0, and can be represented as a stationary object that affects the first travel path L1.

[0076] In step S20, see also Figure 2B and Figure 2EThe host 10 executes the calculation program P and obtains the first travel data L1D by referring to the first travel route L1 of the vehicle V, such as the steering angle and the inner wheel difference. It then performs calculations with the movement data MLD obtained in step S18 to generate a second travel route L2. The host 10 adjusts the first travel data L1D according to the movement data MLD, thereby adjusting the first travel route L1 of the vehicle V and further generating the second travel data L2D of the second travel route L2, such as delaying travel, changing the entry angle of the vehicle V into the parking position 50, and changing the position point of entering the parking position 50. Furthermore, the second travel route L2 generated by the host 10 of the present invention can be displayed on a display unit (not shown) to notify the driver of the vehicle V that he / she needs to avoid dangerous situations caused by blind spots on one side of the vehicle V. The present invention can also be applied to an advanced driver assistance system (ADAS) to allow the ADAS to intervene in the driver's driving behavior, thereby preventing the driver of the vehicle V from avoiding dangerous situations caused by blind spots on one side of the vehicle V.

[0077] The expression for the inner wheel difference is as follows:

[0078]

[0079]

[0080]

[0081] m = ba (Formula 4)

[0082] R is the turning radius of vehicle V, L is the wheelbase, d1 is the front wheel spacing, d2 is the rear wheel spacing, α is the angle between the midpoint of the front and rear axles of vehicle V and the center of the turning circle, a is the radius of motion of the inner rear wheel centerline, b is the radius of motion of the inner front wheel centerline, and m is the inner wheel difference of the non-trailer vehicle.

[0083] like Figure 3 As shown, using perspective projection, the image point P0 captured by the image capturing unit 30 is divided into a first image point P1 and a second image point P2. The coordinates (x, y) of the first image point P1 are located in the first surface region DM1, and the coordinates (x′, y′) of the second image point P2 are located in the second surface region DM2. Therefore, the relative relationship between the image capturing unit 30 in capturing the first image point P1 and the second image point P2 is as follows:

[0084]

[0085]

[0086] Where (x,y) is the first image point P1 and (x',y') is the second image point P2; m0,m1,…m7 are the relevant focal length, rotation angle and scaling parameters of the image capturing unit 30, which can be expanded into a complex array of image point pairs, and then the optimal values ​​of m1 to m7 are obtained by nonlinear minimization operation through the Levenberg-Marquardt algorithm, which are used as the optimal capturing focal length of the image capturing unit 30.

[0087] The aforementioned image optical flow method L uses the Lucas–Kanade Optical Flow algorithm to estimate obstacles. It first obtains the image constraint equations using Taylor's formula through image differencing.

[0088]

[0089] Here, HOT represents a higher-order equation, which can be ignored when the shift is sufficiently small. From this equation, we can obtain:

[0090]

[0091] or

[0092]

[0093] And thus we obtain:

[0094]

[0095] V x V y V z These are the components of x, y, and z in the optical flow vector of I(x,y,z,t). and This is the difference of the image at point (x,y,z,t) in the corresponding direction, so equation (x) is transformed into the following equation.

[0096] I x V x +I y V y +I z V z =-I t Formula (XI)

[0097] Equation (XI) can be further written as follows:

[0098]

[0099] Since there are three unknowns (Vx, Vy, Vz) in equation (10), the unknowns are calculated by the continuation algorithm:

[0100] First assume the flow (V) x V y V z In a small window of size m*m*m (m>1), if the value is a constant, then from pixel 1...n, n=m 3 The following set of equations can be obtained from this:

[0101]

[0102]

[0103] The above equations all contain three unknowns, forming a system of equations. This system is overdetermined, meaning it contains redundancy. The system can be represented as:

[0104]

[0105] Notation:

[0106]

[0107] To solve this overdetermined problem, equation (xv) is obtained using the least squares method:

[0108]

[0109]

[0110] get:

[0111]

[0112] Substitute the result of Equation (XVIII) into Equation (X) to estimate an acceleration vector and a distance information of at least one object, which are used to classify and predict the movement of the object. For example, classify the object image OBJ of the first object VO1 as the filter image IMG, and predict the predicted movement ML of the first object VO1.

[0113] In addition, such as Figure 4As shown, the host 10 can also pre-obtain a first effective area A1 of the parking position 50 and a vehicle size S of the vehicle V, i.e., the visual length and width of the vehicle V. In step S20, the processing unit 12 of the host 10 can further determine whether the first effective area A1 of the parking position 50 has shrunk to a second effective area A2 based on the predicted movement line ML, wherein the first effective area A1 is larger than the vehicle size S of the vehicle V, and the second effective area A2 is smaller than the vehicle size S. When the processing unit 12 of the host 10 determines that the first effective area A1 has shrunk to the second effective area A2, the processing unit 12 adjusts the second travel data L2D so that the second travel route L2 indicates that the vehicle V is parked on part of the parking position 50. For example, a first object VO1 is located on one side edge of the parking position 50, so that the effective area of ​​the parking position 50 is reduced to 80%, and therefore smaller than the size S of the vehicle V. Part of the vehicle V is located on the edge of the parking position 50, or even beyond the edge of the parking position 50. Figure 5 As shown, when the processing unit 12 determines that the effective area of ​​the parking position 50 has not changed, that is, the processing unit 12 maintains the second travel data L2D, so that the second travel route L2 indicates that the vehicle V is parked in the parking position 50.

[0114] In summary, the mobile vehicle lateral blind spot sensing system and method of the present invention provides a host to acquire object images of multiple objects on one side of the vehicle for classification, thereby obtaining a filtered image. The host then performs predictive calculations on the objects mapped from the filtered image to obtain a predicted movement path. This predicted movement path is then calculated against the vehicle's travel data to generate a second travel path. Furthermore, the host can further adjust the travel data based on the movement path data to avoid dangerous situations.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. All equivalent variations and modifications made in accordance with the shape, structure, features and spirit described in the claims of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for lateral blind spot sensing of a mobile vehicle, characterized in that, It is applied to a vehicle equipped with a host computer, an optical scanning unit, and an image capturing unit, wherein the host computer is electrically connected to the optical scanning unit and the image capturing unit. The method includes the following steps: The host executes a parking command by instructing the host to generate a positioning message based on a relative or absolute position of the vehicle relative to the parking position, which is located on one side of the vehicle; The host obtains a first travel route for the vehicle based on the positioning information and a location information corresponding to the parking location, which instructs the vehicle to turn to the parking location; The host computer uses the light scanning unit to scan at least one object corresponding to the parking position according to the first travel route, and the host computer uses the image capturing unit to capture the image of at least one object corresponding to the at least one object. The image of at least one object and the at least one object correspond to the blind spot position on one side of the vehicle. The host computer uses an image optical flow method to filter the at least one object image according to the first travel route, and obtains at least one filtered image corresponding to the at least one object image according to the first travel route. The host generates at least one predicted motion line based on at least one object vector corresponding to at least one screened image; as well as The host modifies the first travel route based on the at least one predicted movement line and generates a second travel route accordingly; In the step of the host modifying the first travel route based on the at least one predicted movement line and generating a corresponding second travel route, the host determines whether a first effective area of ​​the parking position is reduced to a second effective area based on the at least one predicted movement line. The first effective area is larger than the vehicle size, and the second effective area is smaller than the vehicle size. When the first effective area is reduced to the second effective area, the second travel route indicates that the vehicle is parked in part of the parking position.

2. The method for lateral blind spot sensing of a mobile vehicle as described in claim 1, characterized in that, In the step of scanning at least one object corresponding to the parking position using the light scanning unit according to the first travel route and capturing an image of the corresponding at least one object using the image capturing unit, the light scanning unit further scans the at least one object around the parking position and captures an image of the corresponding at least one object around the parking position using the image capturing unit.

3. The method for lateral blind spot sensing of a mobile vehicle as described in claim 1, characterized in that, In the step of classifying the object images using an image optical flow method based on the first travel route and the parking position, the host extracts multiple three-dimensional images based on the filtered images and classifies the object images based on the positioning information and the image optical flow method.

4. The method for lateral blind spot sensing of a mobile vehicle as described in claim 1, characterized in that, In the step of modifying the first travel route and generating a second travel route based on the at least one predicted travel line, the host calculates based on an inner wheel difference and a steering angle corresponding to the first travel route and the at least one predicted travel line, thereby modifying the first travel route and generating the second travel route accordingly.

5. A lateral blind spot sensing system for a mobile vehicle, applied to a vehicle, characterized in that, The mobile vehicle's lateral blind spot sensing system includes: A host computer is installed inside the vehicle. The host computer executes a parking command based on a parking position located on one side of the vehicle. The host computer generates a positioning message based on a relative position or an absolute position of the vehicle relative to the parking position. The host computer obtains a first travel route based on the positioning message and a position message corresponding to the parking position, which is to instruct the vehicle to turn to the parking position. A light scanning unit is disposed on one side of the vehicle and electrically connected to the main unit. The light scanning unit scans at least one object corresponding to the parking position according to the first travel route. The at least one object corresponds to a side blind spot of the vehicle. An image capturing unit is disposed on the side of the vehicle and adjacent to the optical scanning unit. The image capturing unit is electrically connected to the host. The image capturing unit captures an image of at least one object corresponding to at least one object. The image of at least one object corresponds to the lateral blind spot position of the vehicle. The host computer performs an image optical flow method to filter the image of at least one object based on the first travel route to obtain at least one filtered image. The host computer generates at least one predicted movement line based on the at least one object vector of the at least one filtered image. The host computer modifies the first travel route based on the at least one predicted movement line and generates a corresponding second travel route. The host computer determines whether a first effective area of ​​the parking position is reduced to a second effective area based on the at least one predicted movement line. The first effective area is larger than the vehicle size, and the second effective area is smaller than the vehicle size. When the first effective area is reduced to the second effective area, the second travel route indicates that the vehicle is parked in part of the parking position.

6. The lateral blind spot sensing system for mobile vehicles as described in claim 5, characterized in that, The optical scanning unit is either an optical radar device or a laser scanner.

7. The lateral blind spot sensing system for mobile vehicles as described in claim 5, characterized in that, The host calculates based on the inner wheel difference and a steering angle corresponding to the first travel route and the at least one predicted movement line, thereby modifying the first travel route and generating the second travel route accordingly.

8. The lateral blind spot sensing system for mobile vehicles as described in claim 5, characterized in that, The lateral blind spot location is a blind spot area corresponding to the vehicle relative to the parking position and conforming to the specifications in the ISO 17387 standard for intelligent transportation system certification.

Images