Collision avoidance system and collision avoidance method

Abstract

To provide a CAS and collision avoidance method.SOLUTION: A collision avoidance method includes the steps of: detecting a first detection area and a second detection area, which encompasses the first detection area and is larger than the first detection area, through a radar and generating a result of detection of an object; determining based on the result of detection whether the object enters the first detection area or second detection area; calculating a first collision probability on the basis of a first weight and the result of detection in response to the determination that the object enters the second detection area and does not enter the first detection area; calculating the first collision probability on the basis of a second weight and the result of detection in response to the determination that the object enters the first detection area; and determining based on the first collision probability whether to output an alarm message.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to radar systems, and more particularly to a carrier collision avoidance system (CAS) and a collision avoidance method. [Background technology]

[0002] Conventional CAS uses radar or image recognition technology to detect objects approaching the carrier. CAS that relies on radar is prone to misjudgments due to interference from environmental noise, while CAS that uses image recognition technology faces challenges in accurately determining the speed and distance of objects. Therefore, ensuring the provision of reliable and accurate CAS remains a major concern in related fields. Summary of the Invention [Problem to be solved by the invention]

[0003] The present invention provides a CAS and collision avoidance method that utilizes radar and image recognition technology to detect objects approaching the carrier. [Means for solving the problem]

[0004] According to one embodiment of the present invention, a CAS is adapted to a carrier and includes a radar, an output device, and a processor. The processor is coupled to the radar and the output device and is configured to: detect a first detection area and a second detection area including the first detection area but larger than the first detection area through the radar to generate a detection result of a target object; determine whether the target object enters the first detection area or the second detection area based on the detection result; calculate a first collision probability based on the first weight and the detection result in response to determining that the target object enters the second detection area but not the first detection area; calculate the first collision probability based on the second weight and the detection result in response to determining that the target object enters the first detection area; determine whether to output a warning message based on the first collision probability; and output the warning message via the output device in response to determining to output the warning message.

[0005] In one embodiment of the present invention, the CAS further includes an image capture device coupled to a processor, the processor being further configured to: capture images of the first detection area and the second detection area through the image capture device; determine whether the images match the detection result based on a first timestamp of the detection result and a second timestamp of the images; determine, in response to the image matching the detection result, an illumination condition of the target object in the image, and calculate a total collision probability based on the first collision probability and the illumination condition; and determine whether to output a warning message based on the total collision probability.

[0006] In one embodiment of the present invention, the detection result includes a first distance, and the processor is further configured to: determine a second distance corresponding to a target object in the image based on the image; determine whether the second distance matches the first distance; and, in response to the second distance matching the first distance, determine a lighting condition based on the target object in the image.

[0007] In one embodiment of the present invention, the processor is further configured to, in response to determining that the target object enters the second detection area but not the first detection area, calculate a first collision probability based on a first distance between the carrier and the boundary of the first detection area, a second distance between the carrier and the boundary of the second detection area, the detection result, and the first weight.

[0008] According to one embodiment of the present invention, a collision avoidance method is applied to a carrier and includes the following steps: detecting a first detection area for generating a detection result of a target object through a radar, and a second detection area including the first detection area but larger than the first detection area; determining whether the target object will enter the first detection area or the second detection area based on the detection result; calculating a first collision probability based on a first weight and the detection result in response to determining that the target object will enter the second detection area but not the first detection area; calculating a first collision probability based on the second weight and the detection result in response to determining that the target object will enter the first detection area; determining whether to output a warning message based on the first collision probability; and outputting the warning message via an output device in response to determining to output the warning message. [Effects of the Invention]

[0009] In view of the above, the CAS provided in one or more embodiments of the present invention can accurately detect the speed or distance of a target object approaching the carrier, and can thereby determine whether to issue an alarm message. Furthermore, the CAS provided in one or more embodiments of the present invention is designed to withstand interference caused by environmental noise, and thus demonstrates strong reliability.

[0010] To explain the present invention in more detail, several exemplary embodiments accompanied with drawings are described in detail below. [Brief explanation of the drawings]

[0011] The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the invention and, together with the description, serve to explain the principles of the invention. [Figure 1] FIG. 1 is a schematic diagram of a CAS according to one embodiment of the present invention. [Figure 2] FIG. 2 is a schematic diagram showing a configuration of a sensor of a carrier according to an embodiment of the present invention. [Figure 3] 2 is a schematic diagram of a detection area of ​​a radar according to an embodiment of the present invention; [Figure 4] FIG. 2 is a top view of a carrier in motion according to one embodiment of the present invention. [Figure 5] FIG. 2 is a top view of a carrier in motion according to one embodiment of the present invention. [Figure 6] 2 is a flowchart of a collision avoidance method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] 1 is a schematic diagram of a CAS 100 according to one embodiment of the present invention, where the CAS 100 is adapted to detect an object of interest (e.g., object of interest 300 or 400 shown in FIG. 4) approaching a carrier (e.g., carrier 200 shown in FIG. 2). The CAS 100 may include a processor 110, a storage medium 120, one or more radars 130, one or more image capture devices 140, and an output device 150.

[0013] The processor 110 may include, for example, a central processing unit (CPU) or any other programmable general-purpose or special-purpose microcontroller unit (MCU), microprocessor, digital signal processor (DSP), programmable controller, application-specific integrated circuit (ASIC), graphics processing unit (GPU), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (FPGA), or other similar component, or a combination of the aforementioned components. The processor 110 may be coupled to the storage medium 120, the radar 130, the image capture device 140, and the output device 150, and may access and execute multiple modules and applications stored on the storage medium 120. In one embodiment, the processor 110 is communicatively connected to the carrier's electronic control system and may control the carrier. For example, the processor 110 may control the carrier to brake or avoid collisions.

[0014] The storage medium 120 may include, for example, any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid state drive (SSD), similar components, or a combination of the aforementioned components, and the storage medium 120 is configured to store multiple modules or applications that can be executed by the processor 110.

[0015] The radar 130 may be, for example, a pulse compression radar. The detection results obtained through the radar 130 may be applied to determine the distance, relative velocity, or time to collision (TTC) between different objects.

[0016] Image capture device 140 may be, for example, a camera or imaging device configured to capture an image. Image capture device 140 may include an image sensor, such as a complementary metal-oxide semiconductor (CMOS) or a charge-coupled device (CCD). Images captured by image capture device 140 may be used to determine lighting conditions of an object.

[0017] The processor 110 may output a warning message through the output device 150. Here, the warning message serves to warn the driver that an object is approaching the carrier and there is a possibility of a collision between the two. The output device 150 may include, but is not limited to, a light, a display, or a speaker. That is, the output device 150 may output the warning message in the form of a light signal, an image, a sound, etc.

[0018] One or more radars 130 or one or more image capture devices 140 may be separately provided on the carrier to monitor the surroundings of the carrier. Figure 2 is a schematic diagram illustrating a sensor configuration of the carrier 200 according to one embodiment of the present invention. In one embodiment, the radar 130 may include a radar 131 for detecting the front of the carrier 200, a radar 132 for detecting the left front of the carrier 200, a radar 133 for detecting the left side of the carrier 200, a radar 134 for detecting the left rear of the carrier 200, a radar 135 for detecting the rear of the carrier 200, a radar 136 for detecting the right rear of the carrier 200, a radar 137 for detecting the right side of the carrier 200, and a radar 138 for detecting the right front of the carrier 200. Image capture device 140 may include image capture device 141 for monitoring the front of carrier 200, image capture device 142 for monitoring the left side of carrier 200, image capture device 143 for monitoring the rear of carrier 200, and image capture device 144 for monitoring the right side of carrier 200.

[0019] The processor 110 can generate a detection result of a target object by detecting one or more detection areas through the radar. The detection result may indicate the presence or absence of a target object, the distance between the target object and the carrier 200, the relative velocity between the target object and the carrier 200, or the TTC between the target object and the carrier 200. FIG. 3 is a schematic diagram of the detection area of ​​the radar 130 according to an embodiment of the present invention. In one embodiment, the detection area of ​​the radar 130 may include the detection area 10 and the detection area 20. Here, the detection area 20 may include the detection area 10 or may be larger than the detection area 10. The processor 110 can determine whether a target object (e.g., the target object 400) enters the detection area 10 or the detection area 20 based on the detection result obtained through the radar 130. A target object that enters the detection area 10 poses a greater danger to the carrier than a target object that only enters the detection area 20 without entering the detection area 10.

[0020] The processor 110 may be communicatively coupled to an instrument system of the carrier 200 to obtain the hourly speed of the carrier 200. The processor 110 may obtain the detection area 10 or the detection area 20 from a lookup table based on the hourly speed of the carrier 200. The lookup table may be associated with a reaction distance required for a driver to apply the brakes of the carrier 200 or a braking distance required for the brakes to stop the carrier 200. For example, the processor 110 may set the detection area 10 such that the minimum distance between the boundary of the detection area 10 and the center point of the carrier 200 is equal to the reaction distance corresponding to the hourly speed of the carrier 200. The processor 110 sets the detection area 20 such that the minimum distance between the boundary of the detection area 20 and the center point of the carrier 200 is equal to the sum of the reaction distance and the braking distance corresponding to the hourly speed of the carrier 200. [Table 1]

[0021] Table 1 is an example of a lookup table. Assuming that the speed of the carrier 200 is 20 kilometers per hour, the processor 110 can determine, based on the lookup table, that the minimum distance (e.g., distance d1 or distance w1) between the boundary of the detection area 10 and the center point (or chassis) of the carrier 200 is 5.56 meters (i.e., the reaction distance corresponding to 20 kilometers per hour), and that the minimum distance between the boundary of the detection area 20 and the center point (or chassis) of the carrier 200 is 15.56 meters (the sum of the reaction distance and braking distance corresponding to 20 kilometers per hour).

[0022] The detection result of the target object obtained through the radar 130 may indicate whether the target object enters the detection area 20 or the detection area 10. When the target object does not enter the detection area 10 but simply enters the detection area 20, or when the target object enters the detection area 10, the processor 110 calculates the collision probability P area where D is the distance between the target object and the carrier 200, and J 10 and J 20 is the weight and D 20 =w2 (or D 20 =d2), and D 10 =w1 (or D 10 = d1). Table 2 shows the weight J 10 and J 20 The processor 110 calculates the collision probability P based on the contents recorded in Table 2 and the equation (1). area In one embodiment, the processor 110 may calculate the collision probability P area If P is greater than the threshold, it may be determined to output a warning message (or control the carrier 200 to brake or avoid collision), or the collision probability P area If the threshold value is less than or equal to the threshold value, it may be determined that a warning message is not to be output.

number

[0023] The processor 110 can capture the detection area 10 and the detection area 20 through the image capture device 140 and perform image recognition on the image to determine the illumination condition of the target object in the image. The processor 110 can calculate the collision probability P between the target object and the carrier 200 based on the illumination condition of the target object, as shown in equation (2): reaction In the formula, P approaching is a variable associated with the lighting conditions, and TTC is the TTC of the target object obtained by the processor 110 based on the detection result obtained by the radar 130. In one embodiment, if the TTC is less than 2.8 seconds, the processor 110 calculates the collision probability P reaction can be set to 50%. That is, if the TTC is equal to or greater than 2.8 seconds, the processor 110 sets the collision probability P reaction In one embodiment, if processor 110 determines that the lighting conditions of the target object (e.g., tail lights or directional lights) are "valid," processor 110 sets variable P approaching can be set to 25%. If the processor 110 determines that the lighting conditions of the target object are "invalid", the processor 110 sets the variable P approaching In one embodiment, the processor 110 may set the collision probability P reaction If is greater than the threshold, it may be determined to output a warning message (or control the carrier 200 to brake or avoid collision), and the collision probability P reaction If the threshold value is less than or equal to the threshold value, it may be determined that a warning message is not to be output.

number

[0024] In one embodiment, the processor 110 calculates the total collision probability P totalwhere α is a self-defined value (e.g., α=25%), β=P area +P reaction , P approaching =0 indicates that the processor 110 determines the lighting condition of the target object as "invalid", and P approaching ≠0 indicates that the processor 110 determines the lighting condition of the target object as "valid." The processor 110 determines the total collision probability P total is greater than the threshold, it is determined that a warning message should be output (or the carrier 200 should be controlled to brake or avoid collision), and the total collision probability P total If the threshold value is less than the threshold value, it may be determined that a warning message is not to be output.

number

[0025] In one embodiment, the processor 110 can determine, based on the timestamp, whether the image captured by the image capture device 140 matches the detection result obtained through the radar 130. Specifically, the processor 110 can determine whether the timestamp of the detection result matches the timestamp of the image. If the two timestamps match, the processor 110 can determine that the image matches the detection result. If the two timestamps do not match, the processor 110 can determine that the image does not match the detection result. If the image matches the detection result obtained through the radar 130, it indicates that the image can function to compensate for or improve the detection result obtained through the radar 130. Therefore, the processor 110 can perform image recognition on the image to determine the lighting conditions (e.g., directional lights and brake lights) of the target object (e.g., a carrier such as an automobile or motorcycle). If the image does not match the detection result obtained through the radar 130, it indicates that the image cannot compensate for or improve the detection result obtained through the radar 130. Therefore, the processor 110 may not perform image recognition on the image. That is, the processor 110 does not need to determine the lighting conditions of the target object in the image. The matched detection results and the lighting conditions can be applied to calculate the overall collision probability between the target object and the carrier 200.

[0026] In one embodiment, processor 110 can determine whether an image captured by image capture device 140 matches the detection results acquired by radar 130 based on the distances. Specifically, processor 110 can determine a first distance between the target object and carrier 200 from the detection results acquired through radar 130, and can determine a second distance between the target object in the image and carrier 200 from the image captured by image capture device 140. If the first distance matches the second distance (e.g., the absolute difference between the first distance and the second distance is less than or equal to a threshold), this indicates that the target object in the image matches the target object in the detection results acquired through radar 130 (i.e., the two target objects are the same). Thus, processor 110 can perform image recognition on the image to determine the lighting conditions of the target object. If the first distance does not match the second distance (e.g., the absolute difference between the first distance and the second distance is greater than a threshold), it indicates that the object in the image and the object in the detection result obtained through radar 130 do not match (i.e., the two objects are different). Therefore, processor 110 does not need to perform image recognition on the image. That is, processor 110 does not need to determine the lighting conditions of the object in the image. In one embodiment, processor 110 can project the image and the detection result obtained through radar 130 into the same coordinate system and calculate the absolute difference between the first distance and the second distance.

[0027] 4 and 5 are top views of a moving carrier 200 according to one embodiment of the present invention. For example, if the speed of the carrier 200 is 60 kilometers per hour, D 10 is 16.67 km / h, D 20Assuming that the speed of the target object 300 is 46.67 kilometers per hour, the lighting condition of the target object 300 is "invalid" (as shown in FIG. 4), and the speed of the target object 300 is 30 kilometers per hour, the relative speed between the carrier 200 and the target object 300 is |30-60| = 30 km / h = 8.33 m / s, the distance between the carrier 200 and the target object 300 is 30 meters, and the TTC between the carrier 200 and the target object 300 is 3.6 seconds (distance / relative speed = 3.6). Based on the detection result obtained through the radar 130, the processor 110 can determine that the target object 300 has entered the detection area 20 ahead of the carrier 200. Therefore, the processor 110 calculates the collision probability P corresponding to the target object 300 according to equations (1) to (3), as shown in equations (4) to (6). area , collision probability P reaction , or the total collision probability P total can be calculated.

number

[0028] Assuming that α=25% in equation (3), when the lighting condition of the target object 300 is “effective” (e.g., the brake lights are effective as shown in FIG. 5), the collision probability P reaction or total collision probability P total can be updated to the one shown in equations (7)-(8).

number

[0029] For example, if the speed of Carrier 200 is 60 km / h, D 10 is 16.67 km / h, D 20Assuming that the speed of the target object 400 is 46.67 kilometers per hour, the lighting condition of the target object 400 is "invalid" (as shown in FIG. 4). When the speed of the target object 400 is 40 kilometers per hour, the relative speed between the carrier 200 and the target object 300 is |40-60| = 20 km / h = 5.56 m / s, the distance between the carrier 200 and the target object 400 is 3 meters, and the TTC between the carrier 200 and the target object 400 is 0.54 seconds (distance / relative speed = 0.54). Based on the detection result obtained through the radar 130, the processor 110 can determine that the target object 400 has entered the detection area 10 on the right side of the carrier 200. Therefore, the processor 110 calculates the collision probability P corresponding to the target object 400 according to equations (1) to (3), as shown in equations (9) to (11). area , collision probability P reaction , or the total collision probability P total Since the TTC is less than 2.8 seconds, the processor 110 can calculate the collision probability P reaction can be set to 50%. The collision probability P area and the collision probability P reaction Since the sum of P exceeds 100%, the processor 110 total =100%.

number

[0030] Even if the lighting condition of the target object 400 is changed to "valid" (for example, the directional light shown in FIG. 5 is valid), the TTC is less than 2.8 seconds, so the collision probability P reaction remains at 50%, and the collision probability P area and the collision probability P reaction In one embodiment, the sum of the collision probability P area and the collision probability P reaction If the total exceeds 100%, the total is considered to be 100%.

[0031] FIG. 6 is a flowchart of a collision avoidance method according to an embodiment of the present invention. The collision avoidance method can be realized by the CAS 100 shown in FIG. 1. In step S601, a first detection area and a second detection area including the first detection area but larger than the first detection area are detected using a radar, and a target object detection result is generated. In step S602, based on the detection result, it is determined whether the target object will enter the first detection area or the second detection area. In step S603, in response to a determination that the target object will enter the second detection area but not the first detection area, a first collision probability is calculated based on the first weight and the detection result. In step S604, in response to a determination that the target object will enter the first detection area, a first collision probability is calculated based on the second weight and the detection result. In step S605, it is determined whether to output a warning message based on the first collision probability. In step S606, in response to a determination that a warning message will be output, the warning message is output via an output device.

[0032] In summary, the CAS provided in one or more embodiments of the present invention can determine whether to output a warning message when an object approaches the carrier based on a collision probability, thereby alerting the driver of a potential danger. This collision probability may be related to both the driver's reaction time and the carrier's braking distance. Furthermore, the CAS can calculate the collision probability based on the lighting conditions of the object.

[0033] It will be apparent to those skilled in the art that various modifications and variations can be made to the disclosed embodiments without departing from the scope or spirit of the invention. In view of the foregoing, it is intended that the present invention cover modifications and variations provided they come within the scope of the following claims and their equivalents. [Industrial Applicability]

[0034] The collision avoidance system of the present invention can be applied to a vehicle. [Explanation of symbols]

[0035] 10, 20: Detection area 100: Collision Avoidance System 110: Processor 120: Storage medium 130, 131, 132, 133, 134, 135, 136, 137, 138: Radar 140, 141, 142, 143, 144: Image capture device 150: Output device 200: Career 300, 400: Target object S601, S602, S603, S604, S605, S606: Steps

Claims

1. Career fit, Radar and an output device; a processor coupled to the radar and the output device, Detecting a first detection area and a second detection area that includes the first detection area and is larger than the first detection area through the radar, and generating a detection result of a target object; determining whether the target object enters the first detection area or the second detection area based on the detection result; calculating a first collision probability based on a first weight and the detection result in response to a determination that the target object has entered the second detection area but has not entered the first detection area; In response to a determination that the target object will enter the first detection area, calculate the first collision probability based on the second weight and the detection result; determining whether to output a warning message based on the first collision probability; outputting the warning message via the output device in response to the determination to output the warning message; the processor configured to an image capture device coupled to the processor; The processor further comprises: capturing images of the first detection area and the second detection area through the image capture device; determining whether the image matches the detection result based on a first timestamp of the detection result and a second timestamp of the image; responsive to the image matching the detection result, determining a lighting condition of the target object in the image and calculating a total collision probability based on the first collision probability and the lighting condition; determining whether to output the warning message based on the total collision probability; It is configured as follows: Collision avoidance system.

2. The detection result includes a first distance, and the processor further: determining a second distance corresponding to the object in the image based on the image; determining whether the second distance matches the first distance; determining the lighting conditions based on the object in the image in response to the second distance matching the first distance; The collision avoidance system of claim 1 , configured to:

3. The processor further comprises: in response to determining that the target object enters the second detection area but not the first detection area, calculate the first collision probability based on a first distance between the carrier and a boundary of the first detection area, a second distance between the carrier and a boundary of the second detection area, the detection result, and the first weight; It is configured as follows: The collision avoidance system of claim 1 .

4. Career fit, Detecting a first detection area and a second detection area that includes the first detection area and is larger than the first detection area through a radar, and generating a detection result of the target object; determining whether the target object enters the first detection area or the second detection area based on the detection result; calculating a first collision probability based on a first weight and the detection result in response to a determination that the target object has entered the second detection area but has not entered the first detection area; calculating the first collision probability based on the second weight and the detection result in response to determining that the target object will enter the first detection area; determining whether to output a warning message based on the first collision probability; outputting the warning message via an output device in response to determining to output the warning message; capturing images of the first detection area and the second detection area through an image capture device; determining whether the image matches the detection result based on a first timestamp of the detection result and a second timestamp of the image; responsive to the image matching the detection result, determining a lighting condition of the target object in the image and calculating a total collision probability based on the first collision probability and the lighting condition; determining whether to output the warning message based on the total collision probability; A collision avoidance method comprising:

Images