Abnormality detection method, emergency braking control method, domain controller, and computer program product

Abstract

The invention provides an anomaly detection method, an emergency braking control method, a domain controller and a computer program product, and provides an anomaly detection method and an emergency braking control method. The position information corresponding to the road guardrail and the attribute information corresponding to the target object obtained through periodic detection based on environment sensing data are obtained, the environment sensing data at least comprise point cloud data collected based on a radar and image data collected based on a camera, and the attribute information comprises the first position where the target object is located and the target speed; and the second position of the target object after the preset time interval is determined based on the target speed and the first position, if the second position and the first position are located on different sides of the road fence respectively, the target object is determined as an abnormal detection object, vehicle emergency braking control based on the abnormal detection object is avoided, and driving safety is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to an anomaly detection method, an emergency braking control method, a domain controller, and a computer program product. Background Technology

[0002] Automatic Emergency Braking (AEB) is an active safety technology used in vehicles to brake the vehicle in certain emergency situations. AEB primarily uses various sensors to obtain information about other road users, such as the distance and relative speed between the vehicle and other vehicles, pedestrians, or objects ahead. Based on different distances and vehicle speeds, it can determine whether there is a risk of collision. When a potential collision risk is predicted, it actively applies emergency braking or slows down the vehicle, thereby reducing the probability of a collision with the vehicle or pedestrian ahead and avoiding an accident.

[0003] Currently, most automakers employ multi-sensor fusion environmental perception solutions, using not only cameras but also millimeter-wave radar, lidar, and ultrasonic sensors to improve the accuracy and comprehensiveness of information collection. However, due to detection errors, cameras may detect abnormal lateral speeds of targets. When a candidate target exhibits a significant lateral speed, it is easier to trigger the automatic emergency braking system (AEB). False triggering of AEB can lead to unnecessary rear-end collisions or loss of vehicle control, posing driving risks. Summary of the Invention

[0004] Based on this, the present invention provides an anomaly detection method, an emergency braking control method, a domain controller, and a computer program product. By employing this anomaly detection method and emergency braking control method, anomalies are detected in the target object, and anomaly detection objects with detection anomalies are identified, thereby avoiding emergency braking control based on anomaly detection objects, thus improving driving safety and driving experience.

[0005] On one hand, the present invention provides an anomaly detection method, comprising:

[0006] The location information of the road guardrail and the attribute information of the target object are obtained by periodically detecting environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first location of the target object and the target speed.

[0007] Based on the target speed and the first position, determine the second position of the target object after a preset time interval;

[0008] If the second position and the first position are located on different sides of the road fence, then the target object is determined to be an anomaly detection object.

[0009] Furthermore, in some embodiments, the step of detecting the location information corresponding to the road guardrail based on environmental perception data includes:

[0010] Based on the image data, object detection is performed to obtain a first visual detection result corresponding to the road guardrail. The first visual detection result includes the visual detection position corresponding to the road guardrail.

[0011] In the point cloud data, determine a first associated reflection point that is related to the first visual detection result;

[0012] The location information corresponding to the road guardrail is obtained by fusing the location information corresponding to the first associated reflection point and the first visual detection result.

[0013] Furthermore, in some embodiments, the location information includes multiple guardrail coordinate points;

[0014] After determining the second position of the target object after a preset time interval based on the target speed, the method further includes:

[0015] Based on the multiple guardrail coordinate points, determine the guardrail curve equation used to characterize the position of the road guardrail;

[0016] Determine the equation of the moving curve generated by connecting the first position and the second position;

[0017] If the guardrail curve equation and the moving curve equation intersect, then the second position and the first position are determined to be located on different sides of the road fence.

[0018] Furthermore, in some embodiments, the step of periodically detecting the attribute information corresponding to the target object based on environmental perception data includes:

[0019] Based on the image data, object detection is performed to obtain a second visual detection result corresponding to the target object. The second visual detection result includes the visual detection position and visual detection speed corresponding to the target object.

[0020] In the point cloud data, determine a second associated reflection point that is related to the second visual detection result;

[0021] The attribute information corresponding to the target object is obtained by fusing the reflection point information corresponding to the second associated reflection point and the second visual detection result. The reflection point information includes the reflection point position and reflection point velocity corresponding to the second associated reflection point.

[0022] Furthermore, in some embodiments, determining the second position of the target object after a preset time interval based on the target velocity includes:

[0023] Obtain the target speed corresponding to the target object detected from the current period and a preset number of periods before the current period;

[0024] The average speed of the target object is calculated based on the target speed detected in each cycle.

[0025] The second position of the target object after a preset time interval is determined based on the average speed.

[0026] On the other hand, the present invention also provides an emergency braking control method, comprising:

[0027] The detection results are obtained based on environmental perception data, which includes at least point cloud data collected by radar and image data collected by camera. The detection results include at least one target object and attribute information corresponding to each target object.

[0028] Anomaly detection is performed on each target object in the detection result obtained by any of the above anomaly detection methods to determine anomaly detection objects with attribute information detection anomalies.

[0029] Emergency braking control is applied to the vehicle based on the detection results for other target objects besides the abnormal detection objects.

[0030] Furthermore, in some embodiments, the step of performing emergency braking control on the vehicle based on target objects other than the abnormal detection objects in the detection results includes:

[0031] Identify the other target objects besides the abnormal detection objects from the detection results;

[0032] Emergency braking control is performed on the vehicle based on its current vehicle attributes and the attribute information of the other target objects.

[0033] Furthermore, in some embodiments, the current vehicle attributes include at least the vehicle's driving speed, and the attribute information includes the target speed and first position corresponding to the other target objects;

[0034] The emergency braking control of the vehicle based on its current vehicle attributes and the attribute information corresponding to the other target objects includes:

[0035] Determine the radial distance between the vehicle and the other target objects based on the first position;

[0036] Based on the vehicle's driving speed and the target speed, a collision risk assessment is performed. When the collision risk assessment result indicates that there is a collision risk between the vehicle and the other target object, the vehicle is actively braked.

[0037] On the other hand, the present invention also provides a domain controller, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to execute the steps of the above-described anomaly detection method and emergency braking control method.

[0038] On the other hand, the present invention also provides a computer program product having at least one instruction stored thereon, wherein the at least one instruction, when executed by a domain controller, implements the steps of the above-described anomaly detection method and emergency braking control method.

[0039] According to the anomaly detection method and emergency braking control method provided by the present invention, during vehicle operation, the location information of the road guardrail and the attribute information of the target object are acquired through periodic detection based on environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first position and target speed of the target object. Then, based on the target speed and the first position, the second position of the target object after a preset time interval is determined. If the second position and the first position are located on different sides of the road guardrail, the target object is determined to be an anomaly detection object. This method can identify anomaly detection objects in the detection results when the vehicle performs detection based on environmental perception data, thereby avoiding emergency braking control based on anomaly detection objects, thus improving driving safety and driving experience.

[0040] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating an anomaly detection method provided in an embodiment of the present invention.

[0042] Figure 2 A flowchart illustrating an emergency braking control method provided in an embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of the structure of a domain controller provided in an embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0045] In the description of one or more embodiments of the present invention, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0046] Advanced Driver Assistance System (ADAS) is an advanced system integrated into a car to improve driver safety and efficiency. It achieves real-time monitoring and control of the vehicle by integrating a variety of sensors, controllers, and actuators.

[0047] Currently, 1R1V-based fusion environmental perception solutions are widely used in various ADAS systems. This involves using a camera and a millimeter-wave radar to perceive the environment ahead of the vehicle during driving. The data from the camera and radar is fused to improve the accuracy and comprehensiveness of environmental information collection. However, due to detection errors, the camera may detect targets with abnormal lateral speeds. In 1R1V-based fusion environmental perception solutions, the camera's detection results can affect the fusion detection results, meaning the speed of the detected target object may also be abnormal. When a target object is detected with a significant lateral speed, it can easily trigger the automatic emergency braking system (AEB). False triggering of AEB can lead to unnecessary rear-end collisions or loss of vehicle control, posing a significant driving risk.

[0048] Based on this, the present invention proposes an anomaly detection method. During vehicle operation, the method acquires the location information of the road guardrail and the attribute information of the target object obtained by periodically detecting environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first position and speed of the target object. Then, based on the target speed and the first position, the second position of the target object is determined after a preset time interval. If the second position and the first position are located on different sides of the road guardrail, the target object is determined to be an anomaly detection object. This method can identify anomaly detection objects in the detection results when the vehicle performs detection based on environmental perception data, thereby avoiding emergency braking control based on anomaly detection objects, thus improving driving safety and driving experience.

[0049] Environmental perception refers to a vehicle's ability to identify occupants, other vehicles, and the natural environment through sensors mounted inside, outside, and on the road. Vehicle environmental perception is crucial for the accurate operation of various driver assistance functions.

[0050] Please see Figure 2 , Figure 2 This is a flowchart illustrating an anomaly detection method provided in an embodiment of the present invention. The executing entity of this process can be a program for acceleration compensation, or it can be a vehicle or domain controller equipped with the aforementioned program, or other devices capable of communicating with the vehicle, domain controller, etc., without specific limitations.

[0051] The following is about Figure 1 The process shown will be described in detail. The anomaly detection method may specifically include the following steps:

[0052] Step S102: Obtain the location information of the road guardrail and the attribute information of the target object obtained by periodic detection based on environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first location of the target object and the target speed.

[0053] It should be noted that environmental perception refers to the vehicle's ability to identify occupants, vehicles, people, and the road environment through sensors mounted inside, outside, and on the road. In one or more embodiments of the present invention, during vehicle operation, the vehicle performs environmental perception based on radar and cameras, collecting environmental perception data including radar point cloud data and image data, and performs object recognition and detection based on the collected environmental perception data to obtain corresponding detection results.

[0054] In this embodiment of the invention, the detection result obtained based on environmental perception data is acquired. The detection result includes the location information of the road guardrail and the attribute information corresponding to the target object. The attribute information includes the first location of the target object and the target speed.

[0055] Understandably, the anomaly detection method proposed in this invention is mainly applied to road environments containing guardrails. Based on environmental perception data, it can detect various objects in the surrounding environment. The embodiments of this invention primarily obtain the location of the guardrails and the attribute information of the target objects from the detection results. The target objects can be pedestrians, two-wheeled vehicles, animals, or other moving objects that should not be present in the motor vehicle lane.

[0056] Step S104: Determine the second position of the target object after a preset time interval based on the target velocity and the first position;

[0057] Among them, the target speed is the speed information of the target object detected based on environmental perception data, including the speed magnitude and speed direction; the first position is the position coordinate information of the target object in the coordinate system constructed by the vehicle environmental perception system, which is detected based on environmental perception data.

[0058] In this embodiment, based on the first position corresponding to the target object and the target speed corresponding to the target object, the second position of the target object after moving at the target speed for a preset time interval is calculated.

[0059] Preferably, the preset time interval can be 0.8 seconds.

[0060] The preset time interval can also be determined based on the sensitivity of the vehicle's Automatic Emergency Braking (AEB) system or the triggering conditions.

[0061] Step S106: If the second position and the first position are located on different sides of the road fence, then the target object is determined to be an anomaly detection object.

[0062] Specifically, by calculating the second position of the target object after a preset time interval at the target speed, and comparing the positional relationship between the current first position of the target object, the second position after the preset time interval, and the road fence, if the second position and the first position are located on opposite sides of the road fence, the target object detected based on the environmental perception data can be determined as an abnormal detection object, and the target speed corresponding to the target object may be abnormal.

[0063] It is understandable that in roads with guardrails, motor vehicle lanes and non-motor vehicle lanes are separated by the guardrails. Pedestrians, two-wheeled vehicles, and other objects should not appear in the same lane as motor vehicles due to the presence of the guardrails. If a target object detected based on environmental perception data will cross the guardrail within 0.8 seconds according to the detected target speed, then the target speed detection corresponding to the target object is considered abnormal, and the detected target object is determined to be an abnormal detection object.

[0064] According to the anomaly detection method provided by the present invention, during vehicle operation, the method acquires the location information of the road guardrail and the attribute information of the target object obtained by periodically detecting based on environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first position and target speed of the target object. Then, based on the target speed and the first position, the second position of the target object after a preset time interval is determined. If the second position and the first position are located on different sides of the road guardrail, the target object is determined to be an anomaly detection object. This method can identify anomaly detection objects in the detection results when the vehicle performs detection based on environmental perception data, thereby avoiding emergency braking control based on anomaly detection objects, thus improving driving safety and driving experience.

[0065] In one embodiment, when obtaining the location information corresponding to the road guardrail based on environmental perception data, the specific steps may be as follows: object detection is performed based on image data to obtain a first visual detection result corresponding to the road guardrail. The first visual detection result includes the visual detection position corresponding to the road guardrail. Then, a first associated reflection point associated with the first visual detection result is determined in the point cloud data. Finally, the location information corresponding to the first associated reflection point and the first visual detection result are fused to obtain the location information corresponding to the road guardrail.

[0066] It should be noted that the embodiments of the present invention adopt an environmental perception scheme based on radar and camera fusion. In environmental perception, point cloud data is used based on radar and image data is collected based on camera. The point cloud data includes several radar reflection points and attribute information corresponding to each reflection point. The attribute information corresponding to the reflection point includes, but is not limited to, velocity, acceleration, position, etc.

[0067] In this embodiment, after acquiring point cloud data and image data, object recognition and detection are first performed based on the image data to obtain the first visual detection result corresponding to the road fence. The first visual detection result includes the location information and speed information of the road fence. Then, based on the first visual detection result, the point cloud data is filtered to identify each first associated reflection point that is related to the first visual detection result. Specifically, the radar reflection points that are close to the first visual detection result can be determined based on the location and speed of each radar reflection point. The radar reflection points that are close to the first visual detection result are taken as the first associated reflection points. Finally, the location information corresponding to the road fence is obtained by fusing the location information of the first associated reflection points with the visual detection location in the first visual detection result.

[0068] Optionally, determining the first associated reflection point in the point cloud data that is associated with the first visual detection result can be done by: calculating the proximity between the position of each radar reflection point in the point cloud data and the visual detection position based on the visual detection position in the first visual detection result, and taking the radar reflection point whose proximity meets the preset conditions as the first associated reflection point.

[0069] Optionally, the location information corresponding to the road guardrail can be obtained by fusing the location information corresponding to the first associated reflection point and the first visual detection result. This can be achieved by weighted averaging of the location information of each first associated reflection point and the visual detection position in the first visual detection result to obtain the location information corresponding to the road guardrail.

[0070] In one feasible implementation, the environmental perception scheme based on radar and camera fusion can be implemented based on a pre-trained neural network model. Data features of point cloud data and image data are extracted separately, the extracted data features are fused to obtain fused data features, and the fused data features are classified through a fully connected network to obtain the corresponding detection results.

[0071] In one embodiment, the location information includes multiple guardrail coordinate points; then after determining the second position of the target object after a preset time interval based on the target speed in step S104, the method further includes: determining a guardrail curve equation to characterize the position of the road guardrail based on the multiple guardrail coordinate points; determining a moving curve equation generated by connecting the first position and the second position; if the guardrail curve equation and the moving curve equation have an intersection point, then determining that the second position and the first position are located on different sides of the road fence.

[0072] That is, in the detection of road guardrails, the position information of the road guardrails is represented by multiple guardrail coordinate points. When judging the positional relationship between the road guardrails, the first position, and the second position, the guardrail curve equation used to characterize the position of the road guardrails is determined based on the multiple guardrail coordinate points. Then, the movement curve equation of the target object is determined based on the first position and the second position. When the guardrail curve equation and the movement curve equation have an intersection point, it is determined that the second position and the first position are located on different sides of the road guardrail.

[0073] In one embodiment, when obtaining attribute information corresponding to a target object based on environmental perception data, the specific steps may be as follows: performing object detection based on image data to obtain a second visual detection result corresponding to the target object, the second visual detection result including the visual detection position and visual detection speed corresponding to the target object; determining a second associated reflection point in the point cloud data that is associated with the second visual detection result; and fusing the reflection point information corresponding to the second associated reflection point and the second visual detection result to obtain attribute information corresponding to the target object, the reflection point information including the reflection point position and reflection point speed corresponding to the second associated reflection point.

[0074] It should be noted that the embodiments of the present invention adopt an environmental perception scheme based on radar and camera fusion. In environmental perception, point cloud data is used based on radar and image data is collected based on camera. The point cloud data includes several radar reflection points and attribute information corresponding to each reflection point. The attribute information corresponding to the reflection point includes, but is not limited to, velocity, acceleration, position, etc.

[0075] In this embodiment, after acquiring point cloud data and image data, object recognition and detection are first performed based on the image data to obtain a second visual detection result corresponding to the target object. The second visual detection result includes at least the visual detection position and visual detection speed corresponding to the target object. Then, based on the second visual detection result, each second associated reflection point associated with the second visual detection result is filtered in the point cloud data. Specifically, radar reflection points close to the second visual detection result can be determined based on the position and speed of each radar reflection point. The radar reflection points close to the second visual detection result are taken as second associated reflection points. Finally, the attribute information corresponding to the target object is obtained by fusing the reflection point position, reflection point speed, visual detection position and visual detection speed in the second visual detection result.

[0076] Optionally, determining the second associated reflection point in the point cloud data that is related to the second visual detection result can be done by: calculating the proximity between the position of each radar reflection point in the point cloud data and the visual detection position based on the visual detection position in the second visual detection result, and taking the radar reflection point whose proximity meets the preset conditions as the second associated reflection point.

[0077] Optionally, the location information corresponding to the road guardrail can be obtained by fusing the location information corresponding to the second associated reflection point and the second visual detection result. This can be achieved by weighted averaging of the location information of each second associated reflection point and the visual detection position in the second visual detection result to obtain the location information corresponding to the road guardrail.

[0078] In one embodiment, the attribute information corresponding to the target object is obtained by fusing the reflection point information corresponding to the second associated reflection point and the second visual detection result. Specifically, this can be achieved by: averaging the reflection point velocities corresponding to each second associated reflection point to obtain the radar detection velocity, and averaging the reflection point positions corresponding to each second associated reflection point to obtain the radar detection position; weighting the radar detection velocity and the visual detection velocity in the second visual detection result according to a preset weight to obtain the target velocity corresponding to the target object; and weighting the radar detection position and the visual detection position in the second visual detection result according to a preset weight to obtain the first position corresponding to the target object.

[0079] In this embodiment, the radar detection speed and radar detection position are determined by averaging the sums of the second associated reflection points. Then, the radar detection speed and visual detection speed are weighted and summed, and the radar detection position and visual detection position are weighted and summed to obtain attribute information including the target speed and the first position, which can improve the accuracy of the detection of the target object's speed and position.

[0080] In this embodiment, the average speed of the target object in the previous several detection cycles can be calculated based on the target speed detected in the previous several detection cycles, and then the second position of the target object after a preset time interval can be determined based on the average speed.

[0081] Please see Figure 2 , Figure 2 This is a flowchart illustrating an emergency braking control method provided in an embodiment of the present invention. The executing entity of this process can be a program for vehicle control, or it can be a vehicle or domain controller equipped with the aforementioned program, or other devices capable of communicating with the vehicle, domain controller, etc., without specific limitations. The following describes... Figure 2 The process shown will be described in detail. The emergency braking control method may specifically include the following steps:

[0082] Step S202: Obtain the detection results based on environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The detection results include at least one target object and attribute information corresponding to each target object.

[0083] Specifically, during vehicle operation, the vehicle uses radar and cameras to perceive its environment, collecting environmental perception data including radar point cloud data and image data. Based on this data, it identifies and detects objects, obtaining corresponding detection results. These results include at least one target object and its corresponding attribute information.

[0084] It should be noted that the emergency braking control method provided in this embodiment of the invention can be applied to vehicle control during the initiation of emergency braking. The attribute information of the target object can be used as a basis for determining whether the vehicle triggers emergency braking while in motion.

[0085] Step S204: Use the above-described anomaly detection method to perform anomaly detection on each target object in the detection results to determine the anomaly detection objects with attribute information detection anomalies.

[0086] Specifically, after obtaining the detection results based on environmental perception data, before determining whether to perform emergency braking based on the target object in the detection results, the anomaly detection method proposed in the above embodiments is first used to detect the target object and identify the anomaly detection object with abnormal attribute information in the detection results.

[0087] It should be noted that the anomaly detection object proposed in one or more embodiments of the present invention mainly refers to the speed detection anomaly in the attribute information corresponding to the target object.

[0088] Step S206: Based on the detection results, other target objects besides the abnormal detection objects are used to perform emergency braking control on the vehicle.

[0089] Specifically, after identifying abnormal detection objects in the detection results, these abnormal objects are removed. Based on the other target objects in the detection results (excluding the abnormal objects), a decision is made as to whether to trigger emergency braking control. This avoids unnecessary emergency braking when emergency braking control is triggered based on abnormal detection objects, thereby improving driving safety and the driving experience.

[0090] In one embodiment, step S206, which involves performing emergency braking control on the vehicle based on target objects other than the abnormal detection objects in the detection results, can specifically be: determining other target objects other than the abnormal detection objects in the detection results, and then performing emergency braking control on the vehicle based on the current vehicle attributes of the vehicle and the attribute information corresponding to the other target objects.

[0091] In one embodiment, the current vehicle attributes include at least the vehicle's driving speed, and the attribute information includes the target speed and first position corresponding to other target objects. Then, based on the current vehicle attributes and the attribute information corresponding to other target objects, emergency braking control is performed on the vehicle. Specifically, this can be done by: determining the radial distance between the vehicle and other target objects based on the first position, performing a collision risk assessment based on the vehicle's driving speed and the target speed, obtaining a collision risk assessment result, and when the collision risk assessment result indicates that there is a collision risk between the vehicle and other target objects, actively braking the vehicle.

[0092] It should be noted that the emergency braking control method provided in this embodiment of the invention can be applied to an automatic emergency braking system (AEB). An automatic emergency braking system can determine the risk of collision based on different distances and vehicle speeds. When a potential collision risk is predicted, it actively applies emergency braking or slows down the vehicle, thereby reducing the probability of a collision with a pedestrian and preventing an accident.

[0093] In a specific application scenario, during AEB operation, a 1R1V-based fusion environmental perception system periodically collects environmental perception data through cameras and radar. The radar collects point cloud data, and the camera collects image data of the environment ahead. The point cloud data and image data are fused and detected to obtain the detection result corresponding to the current detection cycle. Then, anomaly detection is performed on each target object in the detection result. Specifically, based on the target speed and position of the target object in the detection result, it is predicted whether the target object will cross the road guardrail after a preset time interval to determine whether there is an anomaly. The abnormal detection objects in the detection result are identified, and then emergency braking control is performed on the vehicle based on other target objects in the detection result, thereby avoiding unnecessary emergency braking when emergency braking control is performed based on abnormal detection objects, thus improving driving safety and driving experience.

[0094] In some embodiments, the present invention also provides Figure 3 The diagram shows the structure of a domain controller. Figure 3 At the hardware level, the domain controller includes a processor 11, an internal bus 12, a network interface 13, memory 14, and non-volatile memory 15, and may also include other hardware required for business operations. This domain controller can be installed in a vehicle. The processor 11 can read the corresponding computer program from the non-volatile memory 15 into memory and run it to implement the aforementioned anomaly detection and emergency braking control methods, preventing false triggering of the emergency braking function and improving driving safety.

[0095] In some embodiments, the present invention also provides a computer program product that may store at least one instruction, which may be loaded and executed by a domain processor as described in the above embodiments of the anomaly detection method and emergency braking control method. The specific execution process can be found in the detailed descriptions in the above embodiments, and will not be repeated here.

[0096] Finally, the various embodiments in this invention are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments.

[0097] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. An anomaly detection method, comprising: The location information of the road guardrail and the attribute information of the target object are obtained by periodically detecting environmental perception data. The environmental perception data includes at least point cloud data collected by radar and image data collected by camera. The attribute information includes the first location of the target object and the target speed. Based on the target speed and the first position, determine the second position of the target object after a preset time interval; If the second position and the first position are located on different sides of the road fence, then the target object is determined to be an anomaly detection object.

2. The method according to claim 1, wherein detecting the location information corresponding to the road guardrail based on environmental perception data includes: Based on the image data, object detection is performed to obtain a first visual detection result corresponding to the road guardrail. The first visual detection result includes the visual detection position corresponding to the road guardrail. In the point cloud data, determine a first associated reflection point that is related to the first visual detection result; The location information corresponding to the road guardrail is obtained by fusing the location information corresponding to the first associated reflection point and the first visual detection result.

3. The method according to claim 2, wherein the location information includes multiple guardrail coordinate points; After determining the second position of the target object after a preset time interval based on the target speed, the method further includes: Based on the multiple guardrail coordinate points, determine the guardrail curve equation used to characterize the position of the road guardrail; Determine the equation of the moving curve generated by connecting the first position and the second position; If the guardrail curve equation and the moving curve equation intersect, then the second position and the first position are determined to be located on different sides of the road fence.

4. The method according to claim 1, wherein obtaining the attribute information corresponding to the target object based on periodic detection of environmental perception data includes: Based on the image data, object detection is performed to obtain a second visual detection result corresponding to the target object. The second visual detection result includes the visual detection position and visual detection speed corresponding to the target object. In the point cloud data, determine a second associated reflection point that is related to the second visual detection result; The attribute information corresponding to the target object is obtained by fusing the reflection point information corresponding to the second associated reflection point and the second visual detection result. The reflection point information includes the reflection point position and reflection point velocity corresponding to the second associated reflection point.

5. The method according to claim 4, wherein fusing the reflection point information corresponding to the second associated reflection point and the second visual detection result to obtain the attribute information corresponding to the target object includes: The radar detection velocity is obtained by averaging the reflection point velocities corresponding to each of the second associated reflection points, and the radar detection position is obtained by averaging the reflection point positions corresponding to each of the second associated reflection points. The target speed corresponding to the target object is obtained by weighting and summing the radar detection speed and the visual detection speed in the second visual detection result according to a preset weight. The radar detection position and the visual detection position in the second visual detection result are weighted and summed according to a preset weight to obtain the first position corresponding to the target object.

6. An emergency braking control method, comprising: The detection results are obtained based on environmental perception data, which includes at least point cloud data collected by radar and image data collected by camera. The detection results include at least one target object and attribute information corresponding to each target object. The anomaly detection method described in any one of claims 1 to 5 is used to perform anomaly detection on each target object in the detection result obtained from the detection, and to determine the anomaly detection object with anomaly in attribute information detection. Emergency braking control is applied to the vehicle based on the detection results for other target objects besides the abnormal detection objects.

7. The method according to claim 6, wherein the step of performing emergency braking control on the vehicle based on target objects other than the abnormal detection object in the detection results includes: Identify the other target objects besides the abnormal detection objects from the detection results; Emergency braking control is performed on the vehicle based on its current vehicle attributes and the attribute information of the other target objects.

8. The method according to claim 7, wherein the current vehicle attribute includes at least the vehicle's driving speed, and the attribute information includes the target speed and first position corresponding to the other target objects; The emergency braking control of the vehicle based on its current vehicle attributes and the attribute information corresponding to the other target objects includes: Determine the radial distance between the vehicle and the other target objects based on the first position; Based on the vehicle's driving speed and the target speed, a collision risk assessment is performed. When the collision risk assessment result indicates that there is a collision risk between the vehicle and the other target object, the vehicle is actively braked.

9. A domain controller, comprising: A processor and a memory; wherein the memory stores computer-readable instructions adapted to be loaded by the processor and to perform the steps of the method as claimed in any one of claims 1 to 8.

10. A computer program product having at least one instruction stored thereon, wherein the at least one instruction, when executed by a domain controller, implements the steps of the method according to any one of claims 1 to 8.

Images