TW201818095A - Millimeter wave radar environment identification system for vehicle use for identifying the type of obstacles outside a vehicle and responding instantly - Google Patents

Abstract

The present invention provides a millimeter wave radar environment identification system for vehicle use, which comprises a millimeter wave radar device for detecting obstacles outside the vehicle, and for converting the detected results into reflective information; an energy intensity calculation device, an anti-noise calculation device, and an obstacle width calculation devices respectively in signal connection with the millimeter wave radar device for receiving the reflective information, and respectively calculating the energy intensity information, the signal to noise ratio (SNR), and the width information for transmission to a control device, wherein the control device reintegrates the energy intensity, the SNR, and the width information for identifying the type of the obstacles. This invention is capable of identifying various types of obstacles outside a vehicle to allow the vehicle to respond instantly, for example, when pedestrians are detected, so that corresponding safety and anti-collision measures can be initiated.

Description

Environmental identification system for automotive millimeter wave radar

The invention relates to a vehicle environment recognition system, in particular to a millimeter wave radar vehicle environment recognition system that uses millimeter wave radar to detect obstacles outside the vehicle, and then calculates by applying an algorithm to identify obstacle types.

With the rapid advancement of science and technology, many product technologies required for daily life have also been accompanied by substantial improvements. Especially through the maturity of semiconductor technology, the emergence and evolution of automotive electronic products have been improved, and the vehicles driven by early drivers have been improved. Make today's vehicles not only have the function of "walking", but also pay more attention to many innovative technologies and technologies on the vehicles to further protect the current vehicles and driving, from the early anti-theft system, back-up radar, to today's obstacles. Relevant technologies such as pedestrian recognition, image detection outside the car, and autonomous driving are all conducive to driving safety.

In view of this, drivers should have higher safety and protection when using these advanced automotive electronic technologies, but today's obstacle or pedestrian recognition technology is not easy to completely distinguish pedestrians or other obstacles. For example, please refer to FIG. 1a of the present invention. The general vehicle 10 can detect the pedestrian 12 by detecting the contour of the human body by using the image. However, the image recognition technology is because the camera is installed on the vehicle 10 and the camera has a perspective. The blind spot cannot detect the obstacle image at a close distance. For example, if there is a puppy 14 in front of the vehicle 10, it is easy to ignore and bump into it because of the blind spot of the viewing angle. Continuing, please refer to FIG. 1b of the present invention. Then, a short-range radar detection is added to the vehicle 10 to detect obstacles such as the puppy 14 at a short distance to avoid blind spots in the perspective of driving. With the development of radar detection technology, the width of obstacles can be identified in radar identification, and then it is used to identify obstacles such as pedestrians, bicycles, passenger cars or trucks. Please refer to the first c diagram of the present invention, but in general radar detection, because only the width of the obstacle is used for identification, if the driving vehicle 10 is driving on the road, it is easy to turn the road tree 16 or the street lights and signal lights When the obstacle is misjudged as pedestrian 12, or when vehicle 10 detects pedestrian 12 and road tree 16 at the same time, the radar may misjudge because of the similar width of pedestrian 12 and road tree 16. Once this technology is applied to autonomous driving Or safety and anti-collision measures, it is easy to cause misidentification. In the event of a collision, or when the vehicle 10 cannot recognize the pedestrian 12 or the road tree 16 during automatic driving, it may cause confusion in execution and result in Unexpected results.

Therefore, in view of the lack of identification of general obstacles, the present invention proposes an environment identification system for vehicle millimeter wave radar, which can effectively identify obstacles such as pedestrians, road trees, street lights, or signal lights.

The main object of the present invention is to provide a vehicle millimeter-wave radar environment recognition system, which uses computing devices with different algorithms to identify various obstacles and effectively distinguish obstacles outside the vehicle such as pedestrians, road trees, and street lights. In the implementation of safety anti-collision measures, road trees will be mistaken for pedestrians, leading to the implementation of wrong safety anti-collision measures and affecting the driving safety of passengers and their passengers, completely guaranteeing the driving safety of everyone in the car.

Another object of the present invention is to provide an environment recognition system for a vehicle millimeter wave radar. When driving, the millimeter wave radar can be used to find various obstacles outside the vehicle, so as to avoid the car from colliding with the naked eye and easily overlooked when the vehicle is moving. Obstacles such as obstacles in the blind corner of the driving line of sight, or obstacles very close to the body when reversing and moving forward, millimeter-wave radar can effectively identify obstacles at a close distance to avoid driving Accidentally collided.

In order to achieve the above object, the present invention proposes an environment recognition system for a vehicle millimeter wave radar, which includes a millimeter wave radar device that transmits a millimeter wave signal to the environment outside the vehicle to detect obstacles outside the vehicle. The millimeter wave radar device then The detection result of the millimeter wave signal is converted into a reflection information; an energy intensity calculation device is a signal connected to the millimeter wave radar device to receive the reflection information and obtain the cross-sectional area and power information of the millimeter wave radar device from the reflection information, thereby calculating Obtain the energy intensity information of the obstacle; the primary anti-noise computing device is a signal connected to the millimeter-wave radar device to receive the reflected information and filter the reflected information and noise to calculate the ratio of the signal and noise of the obstacle ; An obstacle width calculation device is a signal connected to a millimeter wave radar device to receive reflection information and self-reflection information to obtain the position information of the millimeter wave radar device and the obstacle, and use these position information to calculate the width information of the obstacle; a control The device is a signal connected energy intensity calculation device, anti-noise calculation device and obstacle The object width calculation device receives the energy intensity information, the ratio value of the signal and the noise, and the width information. The control device integrates the energy intensity information, the ratio value of the signal and the noise, and the width information, and integrates these information to identify the obstacle. The kind of things.

In the present invention, the millimeter wave radar device transmits a millimeter wave signal to the environment outside the vehicle. After the millimeter wave signal is reflected by an obstacle, it is transmitted back to the millimeter wave radar device. The millimeter wave radar device can reflect the Millimeter wave signals are converted into reflected information.

In the present invention, the energy intensity calculation device uses radar cross-section (RCS) and radar information in the reflection information to calculate the energy intensity information of the obstacle, and the anti-noise calculation device uses low-pass filtering to filter the reflections. Information to obtain a signal amplitude, and use high-pass filtering to filter the reflected information to obtain a noise amplitude. The anti-noise computing device then uses the signal amplitude and noise amplitude to calculate the ratio of the signal and noise of the obstacle, and finally the obstacle The width calculation device uses the azimuth of the millimeter wave radar device and the obstacle in the reflection information to calculate the standard deviation of the azimuth, and uses the standard deviation of the azimuth to calculate the width information of the obstacle.

Finally, the present invention further includes an image recognition device signal connection control device. The control device controls the image recognition device and detects obstacles in the environment outside the vehicle. The detected obstacle images are displayed on the image recognition device. .

In the following, detailed descriptions will be made through specific embodiments in conjunction with the accompanying drawings to make it easier to understand the purpose, technical content, features and effects of the present invention.

With the rapid growth of vehicle electronic product technology, and in order to make clear and effective identification of obstacles outside the vehicle in response to vehicle automatic driving or safety anti-collision systems, the present invention has an environment identification for vehicle millimeter wave radar The system is used to solve the shortcomings that the current vehicle radar cannot effectively distinguish the type of obstacles outside the vehicle.

First, please refer to the second figure of the present invention. An environment identification system 20 for a vehicle millimeter wave radar includes a millimeter wave radar device 22, an energy intensity calculation device 24, an anti-noise calculation device 26, and an obstacle width. The computing device 28 and a control device 30. In this embodiment, the control device 30 is an electronic control unit (ECU), and the millimeter-wave radar device 22 includes a millimeter-wave radar capable of transmitting millimeter-wave signals. Among them, the millimeter-wave radar device 22 is connected to an energy intensity calculation device 24, an anti-noise calculation device 26, and an obstacle width calculation device 28. The present invention does not limit the signal connection method to be a wired signal connection or a wireless signal connection. And the millimeter-wave radar device 22, the energy intensity calculation device 24, the anti-noise calculation device 26, and the obstacle width calculation device 28 are connected to the control device 30 by signals.

Next, please refer to the third diagram of the present invention, and continue to refer to the second diagram of the present invention. After explaining the device architecture of the present invention, the operation mode of the environment identification system 20 of the vehicle millimeter wave radar of the present invention will be described next. To further understand how the present invention can effectively distinguish the types of obstacles 32 outside the vehicle. The environment identification system 20 of the vehicle millimeter wave radar of the present invention can be mounted in various vehicles. In this embodiment, a general passenger car 34 is taken as an example first. In order to be able to transmit a millimeter wave signal, the millimeter wave radar device 22 To detect the obstacle 32 in front of the passenger car 34, the millimeter wave radar device 22 is usually mounted at the front position of the passenger car 34 to transmit a millimeter wave signal to the front of the passenger car 34. As for the millimeter wave radar device 22, Where in front of the passenger car 34 or other vehicles depends on the design of the user or the general car manufacturer. The present invention does not limit the installation position. In addition, in this embodiment, the millimeter wave radar device 22 is used first. The signal is connected to the control device 30 as an example. The millimeter-wave radar device 22 is activated through the control device 30 to transmit the millimeter-wave signal to the outside of the passenger car 34. However, the present invention is not limited to this. It is possible to control the device 30 and the millimeter-wave radar device. 22 signals are connected to the vehicle host (not shown), and the vehicle host is used to control the millimeter wave radar device 22 and the control device 30, or when the control device 30 itself is the vehicle owner At the time of the aircraft, the signal of the millimeter wave radar device 22 is connected to the control device 30 as described in the embodiment. The obstacle 32 in this embodiment first uses four different obstacles 32a, 32b, 32c, and 32d as examples. The obstacle 32a is a road tree, the obstacle 32b is a pedestrian, and the obstacle 32c is a signal lamp. The obstacle 32d is a bicycle and its driver, but the number and type of the obstacles 32 in the present invention are not limited. The millimeter-wave radar device 22 transmits millimeter-wave signals to the environment outside the passenger car 34 to detect obstacles 32. As the passenger car 34 moves, obstacles 32a, 32b, 32c, and 32d at different distances can be gradually detected. Among them, the obstacles 32b and 32d are obstacles that may move. When the millimeter-wave signal contacts these obstacles 32a, 32b, 32c, and 32d, part of them will be reflected back to the millimeter-wave radar device 22, and The millimeter-wave radar device 22 can convert the detection result of the reflected millimeter-wave signal into a reflection information. In this embodiment, the millimeter-wave radar device 22 mainly uses a Fast Fourier Transform (FFT) to convert the millimeters after reflection. Wave signals are converted into reflected information.

Following the above paragraph, after converted into reflection information, the millimeter-wave radar device 22 will simultaneously transmit the reflection information to the energy intensity calculation device 24, the anti-noise calculation device 26, and the obstacle width calculation device 28. The energy intensity calculation device 24 receives and The cross-sectional area and power information of the millimeter-wave radar device 22 is obtained from the reflection information to calculate the energy intensity information of the obstacle 32. The anti-noise calculation device 26 receives and filters the signals and noise in the reflection information to calculate The ratio of the signal and noise of the obstacle 32 is obtained. The obstacle width calculation device 28 receives and obtains the position information of the millimeter-wave radar device 22 and the obstacle 32 from the reflection information to calculate the width information of the obstacle 32.

In order to further explain how the energy intensity calculation device 24, the anti-noise calculation device 26, and the obstacle width calculation device 28 calculate the energy intensity information, the ratio of the signal to the noise, and the width information, the energy intensity calculation devices are described in detail below. 24. Calculation method of anti-noise calculation device 26 and obstacle width calculation device 28. The energy intensity calculation device 24 mainly uses the radar cross-section (RCS) and radar information in the converted reflection information, and uses these data to calculate the energy intensity information of the obstacle 32. In this embodiment, the principle of using the radar cross-sectional area and radar power information to calculate the energy intensity information of the obstacle 32 can be calculated by the following formula (1): (1) where Represents the received transmission power of the millimeter-wave radar device 22, Represents the transmission power of the millimeter-wave radar device 22, Represents the antenna gain of the millimeter-wave radar device 22, Represents the distance between the millimeter-wave radar device 22 and the obstacle 32, Radar cross-sectional area representing obstacle 32, Represents the effective area of the receiving antenna, Represents the antenna gain at the receiving end, Representative wavelength, Stands for Pi. In this embodiment, the obstacles 32a, 32b, 32c, and 32d will have energy intensity information in different energy ranges due to different materials, such as the human body, metal signal lights, or trunks of road trees. Can use neural network-like parameter training to classify the energy intensity of different obstacles 32a, 32b, 32c, 32d into different ranges to distinguish different obstacles 32a, 32b, 32c, 32d. There are many types, and the present invention does not limit the energy intensity information of various obstacles 32.

Meanwhile, the anti-noise computing device 26 uses low-pass filtering to filter reflection information to obtain a signal amplitude and high-pass filtering to filter reflection information to obtain a noise amplitude. In this embodiment, the anti-noise computing device 26 may be Low-pass filter (not shown) and high-pass filter (not shown) are provided for low-pass filtering and high-pass filtering. Next, the anti-noise calculation device 26 calculates the ratio of the signal to the noise by using the filtered signal amplitude and the noise amplitude. The calculation principle can be calculated by the following formula (2): (2) where A value representing the ratio of signal to noise. Represents signal amplitude, Represents the noise amplitude. The ratio of the signal to the noise is also related to the energy intensity information. It will have different signal amplitudes and noise amplitudes due to different types of obstacles 32a, 32b, 32c, 32d. Different types of obstacles 32a, 32b, 32c, 32d The reflected signals and noise are filtered to obtain different signal amplitudes and noise amplitudes, and the ratio of the signal to the noise is calculated. At this time, you can also use neural network-like parameter training to convert these different signals. The ratio of the ratio of the signal to the noise is categorized to obtain the range of the ratio of the signal and the noise formed by different obstacles 32a, 32b, 32c, and 32d. The invention does not limit the ratio of the signal to the noise of various obstacles 32.

The obstacle width calculation device 28 also uses the azimuths of the millimeter wave radar device 22 and the obstacle 32 in the reflection information to calculate the standard deviation of the azimuth angle. At this time, the calculation of the standard deviation of the azimuth angle is the same obstacle. The analytical information of 32 in this embodiment mainly uses the standard deviation of the azimuth plus or minus one standard deviation to calculate the width of the obstacle 32. Under normal distribution, 68.2% of the detection points of the obstacle 32 will come from the same reflecting object. The standard deviation of the azimuth angle is used to calculate the width information of the obstacle 32. The calculation principle can be calculated by the following formulas (3), (4), and (5): (3) (4) (5) where Information representing the width of the obstacle 32, Represents the straight line distance from the obstacle, Represents the azimuth of the obstacle, Represents the average of the azimuth, Represents the standard deviation of the azimuth. Different kinds of obstacles 32a, 32b, 32c, 32d may also have different width ranges. Since there are many types of obstacles 32, the present invention does not limit what the width values of various obstacles 32 should be.

Finally, the energy intensity calculation device 24, anti-noise calculation device 26, and obstacle width calculation device 28 transmit the calculated results to the control device 30, and then use the control of the control device 30 to integrate and analyze the energy intensity information. , Signal and noise ratio value and width information to identify the type of obstacle 32 in front of the body of the passenger car 34. For example, when driving the passenger car 34, there are obstacles 32a, 32b, 32c, and 32d in front of the vehicle body. At this time, the energy intensity information of each obstacle 32a, 32b, 32c, and 32d and the ratio of the signal to noise can be detected. The value is used to know whether the obstacles 32a, 32b, 32c, and 32d are human, metal, or plants, and the width information is also used to determine the obstacles 32a, 32b, 32c, and 32d.

The present invention can effectively avoid the loss of obstacle width by using conventional radar detection only. Please refer to the fourth figure, and also refer to the third figure at the same time. For example, when the passenger car 34 has obstacles 32b such as pedestrians, such as When the obstacles 32a of the road tree and the obstacles 32c such as metal signal lights, the width of the recognition may be confused with the obstacles 32a and 32b of similar width. For example, when encountering an obstacle 32a with a width similar to that of a pedestrian, The width W1 of the detected obstacle 32a may be similar to the width W2 of the obstacle 32b, because the width of the pedestrian may be in the range of about 1 meter, and the road tree may also have a range value like the width of the human body of the pedestrian, and In addition, thinner pedestrians may be mistaken for signal lights because they are too narrow. Therefore, the present invention simultaneously uses the energy intensity information, the comparison of the ratio values of the signal and the noise, to identify the types of obstacles of similar width detected, such as metal materials, plants, or the general human body, plus The width calculation can effectively distinguish the kind of obstacle.

Furthermore, as shown in the fifth figure of the present invention, in addition to the millimeter-wave radar environment identification system 20, in addition to the millimeter-wave radar device 22, the energy intensity calculation device 24, the anti-noise calculation device 26, and the obstacle width calculation device, In addition to 28 and a control device 30, an image recognition device 36 may be further included. In this embodiment, the image recognition device 36 may be a combination of a car camera and a car screen, but the invention is not limited thereto. The signal of the image recognition device 36 is connected to the control device 30, so that the environment recognition system 20 of the vehicle millimeter-wave radar can detect obstacles 32 only by using the millimeter wave signal, and can also use the image recognition device 36 to detect the environment outside the vehicle. The obstacle 32 can be displayed on the image recognition device 36 so that the driver can clearly see the obstacle 32 visible to the naked eye through the screen.

The present invention mainly uses three different identification methods: energy intensity information, ratio values of signals and noise, and width information to enhance the discrimination of obstacle types, and does not limit the identification steps, but can also be identified for special needs. The description of the steps can be, for example, first identifying the energy intensity information, the ratio of the signal to the noise, and then identifying the width of the obstacle to confirm whether the obstacle is a pedestrian or other types of identification processes. The above-mentioned types of obstacles are merely exemplary illustrations of the embodiments, and are not limited by the above-mentioned types of obstacles, and the obstacles may also be street lights or various objects visible on the road. In addition to detecting obstacles that are easily overlooked by the naked eye, the invention can more accurately identify different types of obstacles, for example, it can more easily distinguish whether an obstacle is a pedestrian or not, which is conducive to automatic driving or the implementation of safety and anti-collision measures. , Completely protect the safety of drivers and passengers, or passers-by, and avoid unpredictable consequences due to misidentification of obstacles.

The above-mentioned embodiments are only for explaining the technical ideas and characteristics of the present invention. The purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. When the scope of the patent of the present invention cannot be limited, That is, any equivalent changes or modifications made in accordance with the spirit disclosed in the present invention should still be covered by the patent scope of the present invention.

10‧‧‧ Vehicle

12‧‧‧ pedestrian

14‧‧‧ puppy

16‧‧‧ road tree

20‧‧‧Vehicle millimeter wave radar environment identification system

22‧‧‧ millimeter wave radar device

24‧‧‧ Energy intensity calculation device

26‧‧‧Anti-Noise Computing Device

28‧‧‧ obstacle width calculation device

30‧‧‧Control device

32‧‧‧ obstacles

32a, 32b, 32c, 32d

34‧‧‧ minibus

36‧‧‧Image recognition device

W1‧‧‧Width

W2‧‧‧Width

The first diagram a to the first diagram c are schematic diagrams of identifying obstacles by a conventional vehicle. The second figure is a block diagram of an environment identification system for a vehicle millimeter wave radar according to the present invention. The third figure is a schematic diagram of identifying obstacles using a passenger car according to the present invention. The fourth figure is a schematic diagram for identifying the width of an obstacle according to the present invention. The fifth figure is a schematic block diagram of another embodiment of the present invention.

Claims (12)

An environment identification system for a vehicle millimeter wave radar includes: a millimeter wave radar device that uses a millimeter wave signal to the environment outside the vehicle to detect obstacles, and the millimeter wave radar device is a detection result of the millimeter wave signal Converted into reflection information; an energy intensity calculation device connected to the millimeter-wave radar device to receive the reflection information, and obtaining the cross-sectional area and power information detected by the millimeter-wave radar device from the reflection information, and Calculate the energy intensity information of the obstacle; An anti-noise computing device, which is connected to the millimeter-wave radar device to receive the reflection information, and calculate the signal by filtering the signal and noise of the reflection information. Obstacle signal to noise ratio value; an obstacle width calculation device that is connected to the millimeter-wave radar device to receive the reflection information, and learns the millimeter-wave radar device and the obstacle from the reflection information. Position information to calculate the width information of the obstacle; and a control device that is connected to the energy intensity calculation device by a signal The anti-noise calculation device and the obstacle width calculation device receive the energy intensity information, the ratio of the signal to noise, and the width information. The control device integrates the energy intensity information, the signal and noise. The proportional value and the width information to identify the type of the obstacle. The environment identification system for a vehicle millimeter-wave radar according to claim 1, wherein the millimeter-wave radar device transmits the millimeter-wave signal to the environment outside the vehicle so that the millimeter-wave signal is reflected by the obstacle, and then The millimeter-wave radar device is returned, and the millimeter-wave radar device converts the reflected millimeter-wave signal into the reflection information. The environment identification system for a vehicle millimeter wave radar according to claim 2, wherein the millimeter wave radar device converts the reflected millimeter wave signal into the reflection information by a Fast Fourier Transform (FFT). The environmental identification system for a vehicle millimeter wave radar according to claim 1, wherein the control device is an electronic control unit (ECU). The environmental identification system for a vehicle millimeter wave radar according to claim 1, wherein the obstacle is a pedestrian, a vehicle, a road tree, a street lamp, or a signal post. The environment identification system for a vehicle millimeter-wave radar according to claim 1, wherein the energy intensity calculation device uses radar cross-section (RCS) and radar power information in the reflection information to calculate the obstacle Information about the energy intensity of the object. The environmental identification system for a vehicle millimeter wave radar as described in claim 6, wherein the energy intensity information can be expressed as , Where the parameters of the above formula are: On behalf of receiving the transmitted power of the millimeter-wave radar device, Represents the transmitting power of the millimeter-wave radar device, Represents the antenna gain of the millimeter-wave radar device, Represents the distance between the millimeter wave radar device and the obstacle, The radar cross-sectional area representing the obstacle, Represents the effective area of the receiving antenna, Represents the antenna gain at the receiving end, Representative wavelength, Stands for Pi. The environment identification system for a vehicle millimeter wave radar according to claim 1, wherein the anti-noise computing device uses low-pass filtering to filter the reflection information to obtain a signal amplitude, and uses high-pass filtering to filter the reflection information to obtain a Noise amplitude. The anti-noise calculation device uses the signal amplitude and the noise amplitude to calculate the ratio of the signal to the noise of the obstacle. The environmental identification system for a vehicle millimeter wave radar as described in claim 8, wherein the value of the ratio of the signal to the noise can be expressed as , Where the above-mentioned public parameters are: The ratio of the signal to the noise, Represents the signal amplitude, Represents the noise amplitude. The environment identification system for a vehicle millimeter wave radar according to claim 1, wherein the obstacle width calculation device uses the azimuth angle between the millimeter wave radar device and the obstacle in the reflection information to calculate the azimuth angle And the standard deviation of the azimuth angle to calculate the width information of the obstacle. The environment identification system for a vehicle millimeter wave radar as described in claim 10, wherein the width information can be expressed as , , , Where the above-mentioned public parameters are: Represents the width information, Represents the straight line distance from the obstacle, Represents the azimuth, Represents the average of that azimuth, The standard deviation representing the azimuth. The environmental recognition system for a vehicle millimeter wave radar according to claim 1, further comprising an image recognition device, which is connected to the control device to control the image recognition device to detect the environment outside the vehicle by the control device. The obstacle, and the image of the detected obstacle is displayed on the image recognition device.