Multi-sensor obstacle existence probability fusion method and vehicle

Abstract

The invention discloses a multi-sensor obstacle existence probability fusion method and a vehicle, and the method comprises the steps: receiving an obstacle detection result issued after the detection of a plurality of upstream sensors through a fusion module, carrying out the matching association, carrying out the tracking updating of the obstacle existence probability based on a probability model, presetting the false alarm rate and omission rate of the sensors, and carrying out the tracking updating of the obstacle existence probability. Sensor characteristics, environmental factors, obstacle states and distances are comprehensively considered, the obstacle detection and tracking performance of the vehicle in a dynamic scene is improved, the robustness and adaptability are improved, and the real-time performance and the response capability of a self-vehicle system are enhanced.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent driving technology, and specifically relates to a multi-sensor obstacle existence probability fusion method and a vehicle. Background Art

[0002] The perception module in the intelligent driving system is its core component, which is mainly responsible for collecting and understanding the environmental information around the vehicle. Since different types of sensors have their own advantages and limitations, a single sensor is difficult to cope with complex and changing driving scenarios. Therefore, modern intelligent driving systems usually integrate multiple sensors, such as laser radar (LiDAR), millimeter wave radar, camera and ultrasonic sensor.

[0003] With the rapid development of deep learning technology, intelligent driving systems can directly output obstacle detection results based on deep learning networks. However, due to hardware limitations, it is difficult for a single deep learning network to efficiently process all types of sensor information to generate unified obstacle detection results. Therefore, the system usually uses multiple networks to process information from different sensors and then output multiple obstacle detection results. At this time, the fusion module performs result-level fusion processing based on the detection results of multiple sensors, and generates accurate, stable and unified obstacle information through fusion tracking technology.

[0004] The fusion module generally maintains a list of obstacle tracks. The module first matches the detection results of different sensors with the existing tracks in the track list through an association algorithm. If the detection result is associated with an existing track, the track is updated; if it is not associated with any existing track, a new target is initialized and added to the track list. Finally, the result provided by the fusion module to the downstream module is a high-quality track selected from the track list.

[0005] However, not all obstacle tracks can be directly sent to downstream modules. The presence of obstacles in the track does not mean that they actually exist, but may be a false detection by the sensor; the absence of obstacles in the track list does not mean that they do not really exist, but may be a missed detection by the sensor. Therefore, it is necessary to model the existence of obstacles in the track in order to better express existence and manage the life cycle. In traditional methods, the number of tracking times is usually used as an indicator, and the existence of the track is evaluated through a counter; specifically, a counter is set, and each time a track is successfully associated with a certain detection result, the counter increases; conversely, if it fails to associate with a new detection result, the counter decreases. When the counter exceeds the set threshold, it is considered to exist and can be sent downstream, otherwise it is considered to disappear and the track is deleted. In this way, the system can dynamically evaluate and decide whether to send it to the downstream module or delete the track. However, this technical method has the following disadvantages: 1. Sensor differences: Different types of sensors (such as LiDAR, millimeter-wave radar, and cameras) have different detection capabilities and error characteristics. Simply relying on the number of tracking times cannot fully reflect these differences, which may cause some high-quality but low-frequency detection tracks to be underestimated; 2. Environmental impact: Different sensors have different sensitivities to environmental changes. For example, a camera may not perform well at night or in strong light conditions, while a millimeter-wave radar can still maintain good performance in rainy and snowy weather. Counting methods based on fixed thresholds have difficulty adapting to these changes and may affect the accuracy of tracking quality assessment.

[0006] 3. Dynamic scene adaptability: In complex and dynamic driving environments, the movement patterns and frequency of targets may change rapidly. Fixed technical thresholds may not be able to flexibly respond to these changes, resulting in misjudgments or missed detections.

[0007] 4. Uncertainty in the initial stage: Newly initialized tracks may show large uncertainty in the early stage due to insufficient data. Simply relying on the number of tracking times may prematurely trust these tracks, affecting the overall performance of the system. Summary of the invention

[0008] In view of the above problems, the main purpose of the present invention is to design a multi-sensor obstacle existence probability fusion method and vehicle, which integrates the detection results of multiple sensors, manages the life cycle of the obstacles through the modeling and post-fusion process of the obstacle existence probability value and the threshold of the probability value, and solves the problems of sensor differences and environmental factors, as well as adaptability to dynamic scenes.

[0009] In order to achieve the above purpose, the present invention adopts the following technical solutions: A multi-sensor obstacle existence probability fusion method is applied to a vehicle, wherein the vehicle includes multiple sensors for detecting the same area; a fusion module is used to fuse the detection results of multiple sensors and output them; the fusion module internally maintains a set of obstacle track lists, and the track list is defined as a system track; The fusion method comprises the following steps: The fusion module receives obstacle detection results sent down by multiple upstream sensors; Through the association matching algorithm, the detection results of each sensor are associated and matched with the obstacles in the track list; for the detection results that are successfully associated with the obstacles in the system track, the obstacles in the system track are tracked and updated; for the detection results that are not successfully associated with the obstacles in the system track, a new target is initialized and added to the track list; Generate a new system track based on initialization and tracking update, filter within the new system track, and output unified obstacle results under multiple sensors to the downstream.

[0010] As a further description of the present invention, the obstacle detection result sent down by the above sensor after detection includes position and shape; The tracking update of obstacles includes position update, shape update, and existence probability update. The initialization of new targets includes position initialization, shape initialization, and existence probability initialization. The above-mentioned existence probability initialization and existence probability update are performed by constructing a probability model.

[0011] As a further description of the present invention, constructing a probability model includes the following steps: Define the detection result of a single sensor for an obstacle as a binary value : ; Define the real existence of obstacles as a binary value : ; Determine the prior probability: and They are and The prior probability of occurrence, and ; Define the associated probabilities of the test results: False alarm probability ; Missed detection probability ; Probability of detection ; Correct rejection probability ; According to Bayes' theorem, the posterior probability is calculated as follows: ; in, is the prior probability of the real existence of the obstacle, is the probability of the sensor detection result; is the likelihood probability, indicating that the obstacle really exists. In this case, the sensor detects The probability of is the posterior probability, indicating that the current sensor detection result is When the obstacle actually exists probability; when When , the obstacle really exists, and the expression is: ; Combined with Bayes' theorem: ; Taking the likelihood probability into account, we can get the likelihood ratio for: ; According to the ratio of likelihood ratio to prior probability Determine the presence of obstacles; Under multi-sensor, there are Test results , then the likelihood ratio under multi-sensor is: ; ; ; ; in, is the number of sensors that detect the target, For the The probability that a sensor correctly detects an obstacle when the obstacle actually exists, For the The probability that a sensor fails to detect an obstacle when the obstacle actually exists, For the The probability that a sensor falsely detects an obstacle when the obstacle does not actually exist; Taking the logarithm of the above likelihood ratio, we get: ; For each sensor, according to the detection results , define the weight , the expression is: ; but: ; ; Assumptions , then the above is simplified to judge Is it greater than 0? , then the obstacle exists; , then the obstacle does not exist.

[0012] As a further description of the present invention, the obstacle in the system track is set with a probability attribute, and the obstacle probability preset value is set to , the upper limit is set to ; Before the above existence probability initialization and existence probability update, the false alarm rate and missed detection rate of the sensor are preset for different sensors, different types of obstacles, and different environments; For each sensor’s detection result, calculate the weight value of the sensor’s probability update for obstacle presence based on the preset false alarm rate and missed detection rate. , the expression is: .

[0013] As a further description of the present invention, the existence probability initialization includes the following steps: Based on the test results, determine the source and type of the test results; Calculate the logarithmic ratio according to the preset false alarm rate and missed detection rate ; Initialize existence probability , the expression is: ; The calculated Save it to the system track as the existence probability of the new detected target.

[0014] As a further description of the present invention, after the process of the association matching algorithm, for the detection associated with the obstacle in the system track, a detection list is maintained for storing all detections of the obstacle system track.

[0015] As a further description of the present invention, the existence probability update includes the following steps: Obstacles within the system track , define its existence probability before updating as ; Follow the steps below to traverse multiple sensors and process them one by one: S1: For each sensor , based on obstacles Position, speed, shape and sensors Compare the visual range to determine obstacles Whether it is within the sensor FOV; S2: Get sensor The corresponding preset false alarm rate and missed detection rate; S3: If the obstacle In the sensor The obstacle is detected within the FOV and is in the detection list. , then according to the sensor Calculate the weight value based on the preset false alarm rate and missed detection rate , and add it to the total weight; The obstacle is not in the detection list. , then according to the sensor Under the preset false alarm rate and false detection rate, calculate the weight value , and add it to the total weight; S4: If the obstacle In the sensor If it is not detected within the FOV, no calculation is performed and 0 is returned; After traversing all sensors, the accumulated results are compared with the existence probability before the update Add together to get the updated existence probability , the expression is: ; in, is the total number of sensors, To detect obstacles The number of sensors, is the number of items that are not in the corresponding sensor FOV; The above The detection results of each sensor are saved in the detection list.

[0016] As a further description of the present invention, the new system track generation includes the following steps: Set the obstacle probability deletion threshold ; After initialization and tracking update, the existence probability of all obstacles in the system track is traversed. , and compare it with the deletion threshold For comparison, , delete the obstacle; After the traversal is completed, the system track is used as the new system track.

[0017] As a further description of the present invention, the obstacle screening and sending in the new system track includes the following steps: Upper limit based on obstacle existence probability Deletion threshold with existence probability Setting the Threshold , ; The probability of obstacles existing in the new system trajectory With threshold For comparison, When , the existence of the current obstacle is considered reliable, and the information of the obstacle is output to the downstream for processing or decision-making; If the obstacle is found, the user chooses to keep the track of the obstacle but observe it further.

[0018] A vehicle includes multiple sensors and a fusion module for detecting the same area. The fusion module executes the above-mentioned obstacle existence probability fusion method. The obstacle existence probability fusion method is used to track obstacles under multiple sensors.

[0019] Compared with the prior art, the technical effects of the present invention are: The present invention provides a multi-sensor obstacle existence probability fusion method and vehicle, which receives obstacle detection results sent down by multiple upstream sensors through a fusion module, performs matching and association, and then tracks and updates the obstacle existence probability based on a probability model, presets the false alarm rate and missed detection rate of the sensor, comprehensively considers the sensor characteristics, environmental factors, obstacle status and distance, improves the vehicle's obstacle detection and tracking performance in dynamic scenarios, improves robustness and adaptability, and enhances the real-time performance and responsiveness of the vehicle system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the probability fusion method of the present invention; Figure 2 It is a block diagram of information transmission of the present invention; Figure 3 This is a schematic diagram of the existence probability initialization process of the present invention; Figure 4 It is a schematic diagram of the existence probability update process under a single sensor of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings: In one embodiment of the present invention, a multi-sensor obstacle existence probability fusion method is disclosed, referring to Figure 1-4 As shown, for obstacle detection, this embodiment only describes the existence of obstacles, and does not discuss the detailed detection results (such as length, width, height, etc.).

[0022] Specifically, in this embodiment, the method is applied to a vehicle, the vehicle includes multiple sensors for detecting the same area; a fusion module is used to fuse the detection results of the multiple sensors and then output them, the fusion module internally maintains a set of obstacle track lists, and the track list is defined as a system track; The fusion method comprises the following steps: The fusion module receives obstacle detection results sent down by multiple upstream sensors; Through the association matching algorithm, the detection results of each sensor are associated and matched with the obstacles in the track list; for the detection results that are successfully associated with the obstacles in the system track, the obstacles in the system track are tracked and updated; for the detection results that are not successfully associated with the obstacles in the system track, a new target is initialized and added to the track list; Generate a new system track based on initialization and tracking update, filter within the new system track, and output unified obstacle results under multiple sensors to the downstream.

[0023] It should be noted that the above-mentioned association matching algorithm includes but is not limited to Hungarian matching, and also includes other algorithms that can achieve association matching between sensor detection results and obstacles in the track list.

[0024] In this embodiment, the above fusion method is described in detail, as follows: The obstacle detection results sent down by the above-mentioned sensors after detection include but are not limited to position and shape; the tracking update of obstacles includes position update, shape update, and existence probability update, and the initialization of new targets includes position initialization, shape initialization, and existence probability initialization; wherein, the above-mentioned existence probability initialization and existence probability update are performed by constructing a probability model.

[0025] In this embodiment, the fusion detection algorithm based on the maximum a posteriori probability, which constructs a probability model, includes the following steps: Define the detection result of a single sensor for an obstacle as a binary value : ; Define the real existence of obstacles as a binary value : ; Determine the prior probability: and They are and The prior probability of occurrence, and .

[0026] There are four situations that may occur: Two correct cases: True, test result , correctly rejected, there is actually nothing and it is not detected; True, test result 1. Correct detection, there is something, and it is also detected; Two error conditions: True, test result , which is the first type of error: false alarm (there is nothing in reality, but it is detected); True, test result , which is the second type of error: missed detection (something actually exists, but is not detected).

[0027] Define the associated probabilities of the test results: False alarm probability ; Missed detection probability ; Probability of detection ; Correct rejection probability ; Above, according to Bayes' theorem, the posterior probability is calculated, and the expression is: ; in, is the prior probability of the real existence of the obstacle, is the probability of the sensor detection result; is the likelihood probability, indicating that the obstacle really exists. In this case, the sensor detects probability; is the posterior probability, indicating that the current sensor detection result is When the obstacle actually exists probability.

[0028] When judging binary properties such as existence, the larger probability value is usually taken. When , the obstacle is considered to exist, otherwise the obstacle does not actually exist. The expression is: ; Combined with Bayes' theorem formula: ; Taking the likelihood probability into account, we can get the likelihood ratio for: ; According to the ratio of likelihood ratio to prior probability Determine the presence of obstacles.

[0029] In the case of multiple sensors, assuming there are N sensors, then Test results , then the likelihood ratio under multi-sensor is: ; ; ; ; in, is the number of sensors that detect the target, For the The probability that a sensor correctly detects an obstacle when the obstacle actually exists, For the The probability that a sensor fails to detect an obstacle when the obstacle actually exists, For the The probability that a sensor falsely detects an obstacle when it is not actually present, Indicates Sensors; Taking the logarithm of the above likelihood ratio, we get: ; For each sensor, according to the detection results , define the weight , the expression is: ; but: ; Can be changed to: ; It should be noted that since the prior probability of an obstacle is the real existence of the obstacle, it is generally taken as 0.5, that is, there is a 0.5 probability that an obstacle is real and a 0.5 probability that it does not exist. , then the above is simplified to judge Is it greater than 0? , then the obstacle exists; , then the obstacle does not exist.

[0030] Through the above content, each detection result under multiple sensors is specifically modeled as a probability model associated with the false alarm and missed detection of the corresponding sensor , this model is used to better characterize the existence of obstacles.

[0031] In this embodiment, the above existence probability initialization and existence probability update are specifically implemented as follows: Set the existence probability attribute of obstacles in the system track. The default value of the obstacle existence probability is set to , the upper limit is set to .

[0032] The detection results sent from the upstream are multi-dimensional, and the probability of existence only cares about the indicator of whether there is detection. For different detection sources, different types, and different environmental conditions (such as daytime, nighttime, etc.), the corresponding sensor detection false alarm rate and missed detection rate are also different. According to the empirical value, different false alarm rates and missed detection rates are set for different sensors, different types of obstacles, and different environments (which can also be obtained from the indicators of model training).

[0033] Specifically, before the above existence probability update and existence probability initialization, the false alarm rate and missed detection rate of the sensor are preset for different sensors, different types of obstacles, and different environments; For each sensor’s detection result, calculate the weight value of the sensor’s probability update for obstacle presence based on the preset false alarm rate and missed detection rate. , the expression is: .

[0034] In this embodiment, after the process of the association matching algorithm, for the detection that is not associated with any obstacle in the system track, it is necessary to initialize its existence probability. Figure 3 As shown, the existence probability initialization includes the following steps: Based on the test results, determine the source and type of the test results; Calculate the logarithmic ratio according to the preset false alarm rate and missed detection rate ; Initialize existence probability , the expression is: ; The calculated Save it to the system track as the existence probability of the new detected target.

[0035] In this embodiment, after the process of the association matching algorithm, for the detection of obstacles associated with the system track, a detection list is maintained to store all the detections of the obstacle system track. , before the existence probability update is , in total Under the sensor, there are If a sensor detects The sensor detections are saved to a list for subsequent confirmation.

[0036] Specifically, Figure 4 As shown, the above obstacles The existence probability update includes the following steps: Obstacles within the system track , define its existence probability before updating as ; Follow the steps below to traverse multiple sensors and process them one by one: S1: For each sensor , based on obstacles Position, speed, shape and other information and sensors Compare the visual range to determine obstacles Whether it is within the sensor FOV; S2: Get sensor The corresponding preset false alarm rate and missed detection rate; S3: If the obstacle In the sensor If the obstacle is detected within the FOV, it means that the sensor can theoretically observe it. , indicating that the sensor has detected the obstacle normally, then according to the sensor Calculate the weight value based on the preset false alarm rate and missed detection rate , and add it to the total weight; if there is no obstacle in the detection list at this time , indicating that the obstacle is not detected normally, then according to the sensor Under the preset false alarm rate and false detection rate, calculate the weight value , and add it to the total weight; S4: If the obstacle In the sensor If the obstacle is not detected in the FOV, it means that the obstacle is not observable by the sensor. In this case, no calculation is performed and 0 is returned. Will Repeat the above steps for each sensor, and after traversing all sensors, add the accumulated results and the existence probability before the update. Add together to get the updated existence probability , the expression is: ; in, is the total number of sensors, To detect obstacles The number of sensors, is the number of items that are not in the corresponding sensor FOV; The above The detection results of each sensor are saved in the detection list.

[0037] It should be noted that the above existence probability After initialization The existence probability after multiple association updates is the current existence probability of the obstacle.

[0038] In this embodiment, according to the characteristics of the ln function and the false alarm rate and missed detection rate, if an obstacle actually exists, the existence probability increases positively, and if no obstacle exists, the existence probability decreases negatively.

[0039] In this embodiment, the new system track generation includes the following steps: Set the obstacle probability deletion threshold ; After initialization and tracking update, the existence probability of all obstacles in the system track is traversed. , and compare it with the deletion threshold For comparison, , delete the obstacle; After the traversal is completed, the system track is used as the new system track.

[0040] In this embodiment, the obstacle screening and sending in the new system track includes the following steps: Upper limit based on obstacle existence probability Deletion threshold with existence probability Setting the Threshold , ; The probability of obstacles existing in the new system track With threshold For comparison, When , the existence of the current obstacle is considered reliable, and the information of the obstacle is output to the downstream for processing or decision-making; If the obstacle is found, the user chooses to keep the track of the obstacle but observe it further.

[0041] It should be noted that the above threshold The parameters can be adjusted dynamically according to the vehicle status, obstacle status, etc. In addition, when the obstacle is stably tracked, its existence probability cannot increase infinitely, and the upper limit is ; When the obstacle disappears and is not detected, its existence probability cannot be reduced infinitely, and the deletion threshold is set to If the value is less than this value, the system track will be deleted directly.

[0042] A vehicle comprises a plurality of sensors and a fusion module for detecting the same area, wherein the fusion module executes the above-mentioned obstacle existence probability fusion method, and the obstacle is tracked under multiple sensors by the obstacle existence probability fusion method.

[0043] Through the above content, the existence probability fusion method and vehicle of the present invention are disclosed. Compared with the prior art, the present invention has the following advantages: 1. The existence probability fusion method of the present invention adopts the maximum a posteriori probability fusion detection algorithm to design the obstacle existence probability strategy in the fusion process, providing a more accurate obstacle existence assessment; 2. The existence probability fusion method of the present invention relates the obstacle existence probability to the obstacle state, sensor characteristics, etc., and can be subdivided into models, thereby increasing the flexibility and adaptability of the model, improving robustness, and having strong adaptability to dynamic scenarios; 3. The existence probability fusion method of the present invention performs life cycle management on obstacles in the system track during the post-fusion process by setting thresholds, which helps to improve the accuracy and efficiency of obstacle tracking. By setting different release threshold parameters, it can adapt to complex and dynamic driving environments, meet rapidly changing needs, and enhance the real-time and responsiveness of the system. 4. The existence probability fusion method of the present invention fully considers the sensor differences and environmental influences, improves the accuracy of detection, and improves the robustness and adaptability of the system; 5. The existence probability fusion method of the present invention, the false alarm rate and missed detection rate parameter design are obtained from the detection indicators of the upstream model, simplifying the parameter acquisition process.

[0044] The above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in this field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A multi-sensor obstacle presence probability fusion method, characterized by: The method is applied in a vehicle, the vehicle comprising a plurality of sensors for detecting the same area; The fusion module is based on which a fusion module performs fusion of detection results of multiple sensors and outputs the fusion of the detection results. The fusion module internally maintains a set of obstacle track lists, which are defined as system tracks. The fusion method comprises the following steps: The fusion module receives obstacle detection results sent down by multiple upstream sensors; Through the association matching algorithm, the detection results of each sensor are associated and matched with the obstacles in the track list; According to the detection results of obstacles successfully associated with the system track, the obstacles within the system track are tracked and updated; For the detection results that are not successfully associated with obstacles in the system track, a new target is initialized and added to the track list; Generate a new system track based on initialization and tracking update, filter within the new system track, and output unified obstacle results under multiple sensors to the downstream.

2. The multi-sensor obstacle existence probability fusion method according to claim 1, characterized in that: The obstacle detection results sent by the above sensors after detection include position and shape; The tracking update of obstacles includes position update, shape update, and existence probability update. The initialization of new targets includes position initialization, shape initialization, and existence probability initialization. The above-mentioned existence probability initialization and existence probability update are performed by constructing a probability model.

3. The multi-sensor obstacle existence probability fusion method according to claim 2, characterized in that: Building a probability model includes the following steps: Define the detection result of a single sensor for an obstacle as a binary value : ； Define the real existence of an obstacle as a binary value : ； Determine the prior probability: and They are and The prior probability of occurrence, and ; Define the associated probabilities of the test results: False alarm probability ; Missed detection probability ; Probability of detection ; Correct rejection probability ; According to Bayes' theorem, the posterior probability is calculated as follows: ； in, is the prior probability of the real existence of the obstacle, is the probability of the sensor detection result; is the likelihood probability, indicating that the obstacle really exists. In this case, the sensor detects probability; is the posterior probability, indicating that the current sensor detection result is When the obstacle actually exists probability; when When , the obstacle really exists, and the expression is: ； Combined with Bayes' theorem: ； Taking the likelihood probability into account, we can get the likelihood ratio for: ； According to the ratio of likelihood ratio to prior probability Determine the presence of obstacles; Under multi-sensor, there are Test results , then the likelihood ratio under multi-sensor is: ； ； ； ； in, is the number of sensors that detect the target, For the The probability that a sensor correctly detects an obstacle when the obstacle actually exists, For the The probability that a sensor fails to detect an obstacle when the obstacle actually exists, For the The probability that a sensor falsely detects an obstacle when the obstacle does not actually exist; Taking the logarithm of the above likelihood ratio, we get: ； For each sensor, according to the detection results , define the weight , the expression is: ； but: ； ； Assumptions , then the above is simplified to judge Is it greater than 0? , then the obstacle exists; , then the obstacle does not exist.

4. The multi-sensor obstacle existence probability fusion method according to claim 3 is characterized by: The obstacle in the system track is set with a probability attribute. The default value of the obstacle probability is set to , the upper limit is set to ; Before the above existence probability initialization and existence probability update, the false alarm rate and missed detection rate of the sensor are preset for different sensors, different types of obstacles, and different environments; For each sensor’s detection result, calculate the weight value of the sensor’s probability update for obstacle presence based on the preset false alarm rate and missed detection rate. , the expression is: 。 5. The multi-sensor obstacle existence probability fusion method according to claim 4, characterized in that: The existence probability initialization includes the following steps: Based on the test results, determine the source and type of the test results; Calculate the logarithmic ratio according to the preset false alarm rate and missed detection rate ; Initialize existence probability , the expression is: ； The calculated Save it to the system track as the existence probability of the new detected target.

6. A multi-sensor obstacle existence probability fusion method according to claim 4 or 5, characterized in that: After the process of associating the matching algorithm, for the detection associated with the obstacle in the system track, a detection list is maintained to save all the detections of the obstacle system track.

7. The multi-sensor obstacle existence probability fusion method according to claim 6, characterized in that: The existence probability update includes the following steps: Obstacles within the system track , define its existence probability before updating as ; Follow the steps below to traverse multiple sensors and process them one by one: S1: For each sensor , based on obstacles Position, speed, shape and sensors Compare the visual range to determine obstacles Whether it is within the sensor FOV; S2: Get sensor The corresponding preset false alarm rate and missed detection rate; S3: If the obstacle In the sensor The obstacle is detected within the FOV and is in the detection list. , then according to the sensor Under the preset false alarm rate and missed detection rate, calculate the weight value , and add it to the total weight; The obstacle is not in the detection list. , then according to the sensor Under the preset false alarm rate and false detection rate, calculate the weight value , and add it to the total weight; S4: If the obstacle In the sensor If it is not detected within the FOV, no calculation is performed and 0 is returned; After traversing all sensors, the accumulated results are compared with the existence probability before the update Add together to get the updated existence probability , the expression is: ； in, is the total number of sensors, To detect obstacles The number of sensors, is the number of items that are not in the corresponding sensor FOV; The above The detection results of each sensor are saved in the detection list.

8. The multi-sensor obstacle existence probability fusion method according to claim 7, characterized in that: The new system track generation includes the following steps: Set the obstacle probability deletion threshold ; After initialization and tracking update, the existence probability of all obstacles in the system track is traversed. , and compare it with the deletion threshold For comparison, , delete the obstacle; After the traversal is completed, the system track is used as the new system track.

9. The multi-sensor obstacle existence probability fusion method according to claim 8, characterized in that: The obstacle screening and issuance in the new system track includes the following steps: Upper limit based on obstacle existence probability Deletion threshold with existence probability Setting the Threshold , ; The probability of obstacles existing in the new system track With threshold For comparison, When , the existence of the current obstacle is considered reliable, and the information of the obstacle is output to the downstream for processing or decision-making; If the obstacle is found, the user chooses to keep the track of the obstacle but observe it further.

10. A vehicle, characterized in that: The invention comprises a plurality of sensors and a fusion module for detecting the same area, wherein the obstacle existence probability fusion method according to any one of claims 1 to 9 is executed by the fusion module, and the obstacle existence probability fusion method is used to track obstacles under multiple sensors.

Images