Obstacle information reporting method and system

Abstract

The invention discloses a method and a system for reporting obstacle information, which are applied to an FPGA (Field Programmable Gate Array) sending end and comprise the following steps: acquiring original data acquired by a plurality of sensors; preprocessing the original data to obtain target original data; according to the target original data, determining features of an obstacle; classifying and identifying the obstacles according to the features of the obstacles and the target original data to obtain classification information of the obstacles; generating control information based on the features and the classification information of the obstacles, and converting the classification information and the control information of the obstacles into CAN messages; and the CAN message is sent to a CAN controller, and the CAN controller reports the CAN message to a receiving end through a CAN bus. Therefore, the data processing efficiency can be effectively improved by utilizing the parallel processing capability and the optimized processing data of the FPGA transmitting end, and high-speed and reliable data transmission can be further realized through the CAN bus.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for reporting obstacle information. Background Art

[0002] With the rapid development of autonomous driving technology, the ability of vehicles to perceive the surrounding environment has become increasingly critical. Among them, the obstacle detection system is an important part of autonomous driving. It can identify various obstacles on the road in real time and help the vehicle respond accordingly, thereby improving driving safety and driving experience.

[0003] In the existing technology, it mainly relies on a single type of sensor (such as lidar, millimeter wave radar or ultrasonic sensor) to collect information on the road, and processes the collected data through a microprocessor or embedded controller to perform obstacle reporting.

[0004] However, when processing large amounts of sensor data, the microprocessor or embedded controller has a relatively slow computing speed, resulting in a longer system response time. In addition, for vehicles traveling at high speeds, a fast response is crucial. However, the existing data processing speed cannot meet the needs of vehicles traveling at high speeds, so there is an urgent need to improve data processing capabilities. Summary of the invention

[0005] Based on the above-mentioned deficiencies of the prior art, the present application provides a method and system for reporting obstacle information to solve the problem of low data processing efficiency brought about by the prior art.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] The first aspect of the present application provides a method for reporting obstacle information, which is applied to an FPGA transmitting end, comprising:

[0008] Get the raw data collected by multiple sensors;

[0009] Preprocessing the original data to obtain target original data;

[0010] Determining the characteristics of the obstacle based on the target original data;

[0011] Classifying and identifying the obstacle according to the characteristics of the obstacle and the target original data to obtain classification information of the obstacle;

[0012] generating control information based on the characteristics of the obstacle and the classification information, and converting the obstacle classification information and the control information into a CAN message;

[0013] The CAN message is sent to a CAN controller, and the CAN controller reports the CAN message to a receiving end via a CAN bus.

[0014] Optionally, in the above-mentioned obstacle information reporting method, the preprocessing of the original data to obtain target original data includes:

[0015] Performing filtering on the original data to obtain filtered original data;

[0016] The filtered original data is subjected to noise reduction processing to obtain target original data.

[0017] Optionally, in the above-mentioned method for reporting obstacle information, determining the characteristics of the obstacle according to the target original data includes:

[0018] Inputting the target raw data into a Kalman filter for feature prediction, and predicting the position and relative speed of the obstacle;

[0019] The position and relative speed of the obstacle are determined as characteristics of the obstacle.

[0020] Optionally, in the above-mentioned obstacle information reporting method, classifying and identifying the obstacle according to the characteristics of the obstacle and the target original data to obtain the classification information of the obstacle includes:

[0021] Extracting basic information of the obstacle from the target raw data;

[0022] The basic information and the features of the obstacle are input into a pre-trained machine learning model, and the classification information of the obstacle is output.

[0023] Optionally, in the above-mentioned obstacle information reporting method, the generating control information based on the obstacle characteristics and the classification information includes:

[0024] generating a speed command according to a relative speed in a characteristic of the obstacle;

[0025] Calculating a steering angle of the vehicle according to the obstacle classification information and the position of the feature of the obstacle, and generating a steering instruction based on the steering angle;

[0026] Control information is generated according to the speed command and the steering command.

[0027] The first aspect of the present application provides a reporting system for obstacle information, which is applied to an FPGA transmitting end and includes:

[0028] A data acquisition unit, used to acquire raw data collected by multiple sensors;

[0029] A preprocessing unit, used for preprocessing the original data to obtain target original data;

[0030] A feature determination unit, used to determine the features of the obstacle based on the target original data;

[0031] A classification unit, configured to classify and identify the obstacle according to the characteristics of the obstacle and the target original data, and obtain classification information of the obstacle;

[0032] an information generating unit, configured to generate control information based on the characteristics of the obstacle and the classification information, and convert the obstacle classification information and the control information into a CAN message;

[0033] The reporting unit is used to send the CAN message to the CAN controller, and the CAN controller reports the CAN message to the receiving end through the CAN bus.

[0034] Optionally, in the above obstacle information reporting system, the pre-processing unit includes:

[0035] A filtering processing unit, used for filtering the original data to obtain filtered original data;

[0036] The noise reduction processing unit is used to perform noise reduction processing on the filtered original data to obtain target original data.

[0037] Optionally, in the above obstacle information reporting system, the feature determination unit includes:

[0038] A prediction unit, used for inputting the target raw data into a Kalman filter for feature prediction, and predicting the position and relative speed of the obstacle;

[0039] The feature sub-determination unit is used to determine the position and relative speed of the obstacle as the features of the obstacle.

[0040] Optionally, in the above obstacle information reporting system, the classification unit includes:

[0041] An extraction unit, used to extract basic information of the obstacle from the target raw data;

[0042] An input unit is used to input the basic information and the characteristics of the obstacle into a pre-trained machine learning model and output the classification information of the obstacle.

[0043] Optionally, in the above obstacle information reporting system, the information generating unit includes:

[0044] A first generating unit, configured to generate a speed instruction according to a relative speed in a characteristic of the obstacle;

[0045] a second generating unit, configured to calculate a steering angle of the vehicle according to the obstacle classification information and a position in the feature of the obstacle, and to generate a steering instruction based on the steering angle;

[0046] The third generating unit is used to generate control information according to the speed instruction and the steering instruction.

[0047] The present application provides a method for reporting obstacle information, which is applied to the FPGA transmitter, by obtaining the raw data collected by multiple sensors, and then pre-processing the raw data to obtain the target raw data, then determining the characteristics of the obstacle according to the target raw data, and then classifying and identifying the obstacle according to the characteristics of the obstacle and the target raw data to obtain the classification information of the obstacle, and then generating control information based on the characteristics and classification information of the obstacle, and converting the obstacle classification information and control information into CAN messages, and finally sending the CAN messages to the CAN controller, which reports the CAN messages to the receiving end through the CAN bus. Thus, by utilizing the parallel processing capability and optimized processing data of the FPGA transmitter, the data processing efficiency can be effectively improved, and through the CAN bus, high-speed and reliable data transmission can be further achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0049] Figure 1 A schematic diagram of the structure of a system for reporting obstacle information provided in an embodiment of the present application;

[0050] Figure 2 A flowchart of a method for reporting obstacle information provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of a flow chart of a method for preprocessing raw data provided in another embodiment of the present application;

[0052] Figure 4 A flowchart of a method for determining obstacle characteristics provided in another embodiment of the present application;

[0053] Figure 5A flowchart of an obstacle classification method provided in another embodiment of the present application;

[0054] Figure 6 A flowchart of a method for generating control information provided by another embodiment of the present application;

[0055] Figure 7 A structural diagram of a system for reporting obstacle information provided in another embodiment of the present application. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0057] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0058] The embodiment of the present application provides a method for reporting obstacle information, which is applied to a reporting system for obstacle information to solve the problem of low data processing efficiency brought about by the prior art.

[0059] Alternatively, if Figure 1 As shown, an embodiment of the present application provides a system for reporting obstacle information, including: a radar module, an FPGA main control module and a communication layer.

[0060] Among them, the radar module uses high-frequency millimeter-wave radar with high resolution and long-distance detection capabilities, which is used to scan the surrounding environment in real time and obtain information such as the location and speed of obstacles.

[0061] The FPGA main control module uses a high-performance FPGA chip provided by the Fudan Microelectronics platform, with a built-in custom hardware accelerator to support complex algorithm operations and is used to process a series of obstacle information.

[0062] The communication layer is configured as a high-speed CAN interface that complies with the ISO 11898 standard, with a maximum transmission rate of up to 1Mbps, which is used to transmit obstacle information.

[0063] Based on the reporting system provided above, the present application embodiment provides a method for reporting obstacle information, such as Figure 2 As shown, it is applied to the FPGA sending end and specifically includes the following steps:

[0064] S201. Acquire raw data collected by multiple sensors.

[0065] It is understandable that the surrounding environment can be scanned in real time by a high-frequency millimeter radar to obtain the raw data of the surrounding environment, or the raw data of the surrounding environment collected by various sensors can be received. Optionally, the reading operation of each sensor can be triggered regularly by a timer to ensure that the collection time of each sensor is consistent, thereby synchronizing the information. Among them, the raw data refers to point cloud data one by one, which can be used to characterize the basic information of obstacles.

[0066] S202: Preprocess the original data to obtain target original data.

[0067] Specifically, in order to clean the data, remove unnecessary noise and outliers, and enhance the signal quality of the data, the raw data needs to be preprocessed.

[0068] Optionally, in another embodiment of the present application, a specific implementation of step S202 is as follows: Figure 3 As shown, the following steps are included:

[0069] S301 , filter the original data to obtain filtered original data.

[0070] It is understandable that in order to remove noise from the data, smooth signals, and highlight meaningful trends or patterns, the raw data can be filtered, because filtering can effectively reduce high-frequency noise and interference signals, thereby making the data more stable and reliable, so that after the filtering operation, clean data with noise and unnecessary components removed is obtained.

[0071] Optionally, a suitable filtering algorithm can be selected according to the data characteristics and noise type. Common filtering methods include: low-pass filter (used to remove high-frequency noise), high-pass filter (used to remove low-frequency noise) and band-pass filter (used to retain signals in a specific frequency band), etc.

[0072] S302: Perform noise reduction processing on the filtered original data to obtain target original data.

[0073] Specifically, after filtering the original data, it is also necessary to perform noise reduction processing on the original data to remove the noise in the data, making the signal purer, thereby improving the processing efficiency of subsequent data.

[0074] S203: Determine the characteristics of the obstacle according to the original target data.

[0075] It should be noted that in a dynamic environment, the motion characteristics of obstacles are very important, especially in autonomous driving and robot navigation. According to the changes in the target raw data, dynamic features can be extracted in the following ways:

[0076] If the data comes from a continuous time series (such as video frames or sensor data streams), the movement trajectory of obstacles can be tracked using object tracking algorithms (such as Kalman filtering, particle filtering, or optical flow methods).

[0077] If an object does not change significantly in consecutive frames, it may be a static obstacle, while if the object position or shape changes, it may be a dynamic obstacle (such as a pedestrian or vehicle).

[0078] If the target raw data comes from sensors such as radar, LiDAR, or sonar, they provide reflection intensity and other relevant information about obstacles. For radar or LiDAR sensors, the reflection intensity (ReturnIntensity) can help identify the material of the obstacle, the smoothness of the surface, and its distance from the sensor. Strong reflection signals may indicate hard obstacles (such as buildings or metal objects), while weaker reflections may indicate soft obstacles (such as trees or people).

[0079] Optionally, the distance information measured by a distance sensor (such as a lidar or ultrasonic sensor) can be used to estimate the relative position between the obstacle and the sensor. Thus, by fusing multiple sensor data, the precise position and angle of the obstacle can be calculated, helping the system understand the spatial layout of the obstacle.

[0080] Optionally, in another embodiment of the present application, a specific implementation of step S203 is as follows: Figure 4 As shown, the following steps are included:

[0081] S401, inputting the original target data into the Kalman filter for feature prediction, and predicting the position and relative speed of the obstacle.

[0082] It should be noted that information such as the speed, acceleration and direction of movement of the obstacle is a very important dynamic feature, which can be used to infer the behavior of the obstacle or predict its future position, thereby helping the vehicle to make better decisions and avoid accidents. Therefore, in an embodiment of the present application, the target raw data can be input into the Kalman filter for feature prediction. The Kalman filter can help us estimate the position and relative speed of the obstacle, especially in a noisy environment. It can make estimates based on the dynamic model of the system (such as the motion equation of the obstacle) and the observation data of the sensor, and continuously update the position, speed and other features of the obstacle.

[0083] S402: Determine the position and relative speed of the obstacle as the characteristics of the obstacle.

[0084] S204: classify and identify the obstacles according to the characteristics of the obstacles and the original target data, and obtain classification information of the obstacles.

[0085] It is understandable that in order to improve the intelligence and accuracy of the system, especially in the fields of autonomous driving, robot navigation or environmental perception, it is necessary to know the type of obstacles (such as pedestrians, vehicles or other objects), so it is necessary to classify the obstacles based on the currently known data and features.

[0086] Specifically, obstacle classification and recognition can help the system understand the surrounding environment and make intelligent decisions. For example, autonomous vehicles need to identify and classify different types of obstacles on the road, such as pedestrians, vehicles, and roadblocks, in real time. In this way, the system can take different actions (such as obstacle avoidance, deceleration, parking, etc.) according to different types of obstacles. Once the system identifies and classifies the surrounding obstacles, more efficient path planning can be carried out. For example, the system can adopt different obstacle avoidance strategies for dynamic obstacles (such as pedestrians and vehicles) and static obstacles (such as road signs, buildings, etc.). If it is a mobile obstacle, the system may need to predict its trajectory. If it is a static obstacle, the system only needs to avoid it.

[0087] Optionally, in another embodiment of the present application, a specific implementation of step S204 is as follows: Figure 5 As shown, the following steps are included:

[0088] S501. Extract basic information of obstacles from target raw data.

[0089] The basic information of the obstacle may include shape, size, and motion state (static or dynamic).

[0090] S502: Input the basic information and features of the obstacle into a pre-trained machine learning model, and output the classification information of the obstacle.

[0091] Specifically, the characteristics and basic information of obstacles are obtained through machine learning models, and then the information required for obstacle classification, such as shape, size, etc., is extracted from the basic information. Then, different types of obstacles are distinguished based on the extracted information and characteristics, thereby obtaining the classification information of the obstacles.

[0092] Optionally, the machine learning model is pre-trained using a large amount of obstacle type information (e.g., pedestrians, vehicles, animals, static obstacles, etc.).

[0093] S205 : Generate control information based on the characteristics and classification information of the obstacle, and convert the obstacle classification information and the control information into a CAN message.

[0094] It should be noted that in order to achieve intelligent response to the environment in systems such as autonomous driving and robot navigation, and to ensure that the system can make appropriate decisions based on the category, location, motion state and other characteristics of the obstacles to ensure safety, path optimization and task completion, it is necessary to generate control information based on the characteristics and classification information of the obstacles.

[0095] It should also be noted that since the communication layer of this system is the CAN bus communication interface, in order for the receiving end to receive the control information and make corresponding responses, it is necessary to convert the obstacle classification information and control information into CAN messages in advance to achieve efficient data transmission.

[0096] Optionally, in another embodiment of the present application, a specific implementation method of generating control information based on the characteristics of the obstacle and the obstacle classification information in step S205 is as follows: Figure 6 As shown, the following steps are included:

[0097] S601: Generate a speed instruction according to the relative speed in the characteristics of the obstacle.

[0098] Specifically, if the relative speed is positive (the obstacle is moving away from the target object): If the relative speed between the obstacle and the target object is positive (i.e., the obstacle is moving away from the target object), the target object can maintain the current speed or continue to accelerate, and the speed instruction is generated to continue to accelerate or maintain the current speed. If the relative distance between the target object and the obstacle is large enough and there is no other danger, the speed instruction can be normal acceleration or maintaining the current speed.

[0099] If the relative speed between the obstacle and the target object is negative (i.e. the obstacle is approaching the target object), the system must evaluate whether it needs to slow down, stop or avoid, and generate a speed instruction that if the obstacle approaches quickly, the system should immediately slow down or even make an emergency stop. If the obstacle approaches slowly, the system can avoid a collision by slowing down and planning a new path.

[0100] If the relative speed between the obstacle and the target object is zero, it means that the speed of the obstacle is the same as that of the target object, and the relative position of the target object and the obstacle remains unchanged. The speed command generated is In this case, the system needs to evaluate whether it needs to continue driving or avoid. If the obstacle is dynamic (such as a moving vehicle), the system may need to determine whether a collision will occur. If the obstacle is a stationary object, a detour may need to be planned.

[0101] Optionally, the system needs to not only determine the current relative speed, but also predict the future behavior of the obstacle. Based on the acceleration or speed change trend of the obstacle, the system can determine whether the obstacle is likely to accelerate or decelerate, and generate a speed instruction. If the relative speed of the obstacle is increasing (the approach speed is increasing), the system may need to respond in advance, increase the deceleration amplitude or choose an avoidance path. If the obstacle is decelerating (the speed away from the target object is increasing), it can be accelerated appropriately.

[0102] S602: Calculate the steering angle of the vehicle according to the obstacle classification information and the position of the obstacle's features, and generate a steering instruction based on the steering angle.

[0103] Specifically, the position in the feature of the obstacle refers to the relative position between the obstacle and the vehicle (front, rear, left, right, or relative angle in front of the vehicle).

[0104] It should be noted that the purpose of calculating the steering angle is to ensure that the vehicle can adjust its direction to bypass or avoid obstacles. If the obstacle is in the vehicle's path, steering is necessary. The specific steering angle will be calculated based on the distance and position of the obstacle to the vehicle. If the obstacle is in front of the vehicle, the steering angle should be adjusted to the left or right so that the vehicle can bypass the obstacle. If the obstacle is on the left, the vehicle will turn right. Conversely, if the obstacle is on the right, the vehicle will turn left. If the obstacle is close to the side of the vehicle, more steering angle adjustment is required to avoid a collision.

[0105] Calculate the required steering angle:

[0106] Based on the relative positions of the vehicle and the obstacle, trigonometric functions can be used to calculate the vehicle's steering angle.

[0107] Assuming that the relative position of the obstacle is (x, y) and the current position of the vehicle is the origin (0, 0), the steering angle (θ) can be approximately calculated by the following formula: θ = atan2(y, x).

[0108] Among them, atan2(y, x) is a polar coordinate calculation method based on the (x, y) point. It returns the angle between the point and the horizontal line and is usually used to calculate the angle of rotation.

[0109] If the obstacle is far from the vehicle, the steering angle may be smaller, and if the obstacle is close, the steering angle will be increased.

[0110] It should also be noted that during vehicle driving, there is a certain relationship between the steering angle and the vehicle speed. If the vehicle speed is fast, a larger steering angle may cause loss of control or large deviation, so the steering angle should be limited, usually between 5-20 degrees.

[0111] Formula reference: The steering angle (θ) may be inversely proportional to the vehicle's current speed (v) (to avoid sharp turns that may lead to loss of control).

[0112] In addition, based on the calculated angle (θ), a steering command suitable for the current driving situation is generated. Typically, the steering command will be an output signal proportional to the steering angle.

[0113] The control system can convert the steering angle θ into a specific steering angle command. The output of the specific steering angle may be:

[0114] If θ>0, it means turning right.

[0115] If θ<0, it means turning left.

[0116] If θ is close to 0, it means the vehicle can continue to move forward or maintain the current path.

[0117] The steering command can be a numerical value, such as:

[0118] Steering command = θ, indicating that the vehicle needs to turn at an angle θ.

[0119] S603: Generate control information according to the speed instruction and the steering instruction.

[0120] S206: Send the CAN message to the CAN controller, and the CAN controller reports the CAN message to the receiving end through the CAN bus.

[0121] It is understandable that the FPGA transmitter will send the converted CAN message to the CAN controller, which is responsible for sending the CAN message to the receiver (ECU) through the CAN bus, such as other vehicle systems (such as electronic control unit ECU). When the receiver receives the CAN message, it will parse the CAN message and then take corresponding measures according to the control information in the CAN message. For example, if a stationary vehicle is detected 50 meters ahead, the ECU will first issue a deceleration command according to the command in the control information to gradually decelerate the vehicle to ensure that there is enough time to evaluate the situation and make further responses. At the same time, the system will activate the sound alarm in the car, etc., to remind the user that there is a dangerous situation ahead.

[0122] The present application provides a method for reporting obstacle information, which is applied to the FPGA transmitter, by obtaining the raw data collected by multiple sensors, and then pre-processing the raw data to obtain the target raw data, then determining the characteristics of the obstacle according to the target raw data, and then classifying and identifying the obstacle according to the characteristics of the obstacle and the target raw data to obtain the classification information of the obstacle, and then generating control information based on the characteristics and classification information of the obstacle, and converting the obstacle classification information and control information into CAN messages, and finally sending the CAN messages to the CAN controller, which reports the CAN messages to the receiving end through the CAN bus. Thus, by utilizing the parallel processing capability and optimized processing data of the FPGA transmitter, the data processing efficiency can be effectively improved, and through the CAN bus, high-speed and reliable data transmission can be further achieved.

[0123] Another embodiment of the present application provides a reporting system for obstacle information, such as Figure 7 As shown, it includes the following units:

[0124] The data acquisition unit 701 is used to acquire raw data collected by multiple sensors.

[0125] The preprocessing unit 702 is used to preprocess the original data to obtain target original data.

[0126] The feature determination unit 703 is used to determine the features of the obstacle according to the target original data.

[0127] The classification unit 704 is used to classify and identify obstacles according to the characteristics of the obstacles and the original target data to obtain classification information of the obstacles.

[0128] The information generating unit 705 is used to generate control information based on the characteristics and classification information of the obstacle, and convert the obstacle classification information and the control information into a CAN message.

[0129] The reporting unit 706 is used to send the CAN message to the CAN controller, and the CAN controller reports the CAN message to the receiving end through the CAN bus.

[0130] It should be noted that the specific working process of the above modules in the embodiment of the present application can refer to steps S201 to S206 in the above method embodiment, which will not be repeated here.

[0131] Optionally, in a reporting system for obstacle information provided by another embodiment of the present application, the preprocessing unit 702 includes:

[0132] The filtering processing unit is used to perform filtering processing on the original data to obtain filtered original data.

[0133] The noise reduction processing unit is used to perform noise reduction processing on the filtered original data to obtain target original data.

[0134] Optionally, in a reporting system for obstacle information provided by another embodiment of the present application, the feature determination unit 703 includes:

[0135] The prediction unit is used to input the original target data into the Kalman filter for feature prediction, and predict the position and relative speed of the obstacle.

[0136] The feature sub-determination unit is used to determine the position and relative speed of the obstacle as the features of the obstacle.

[0137] Optionally, in a reporting system for obstacle information provided by another embodiment of the present application, the classification unit 704 includes:

[0138] The extraction unit is used to extract basic information of obstacles from the target raw data.

[0139] The input unit is used to input basic information and features of obstacles into a pre-trained machine learning model and output classification information of obstacles.

[0140] Optionally, in a reporting system for obstacle information provided by another embodiment of the present application, the information generating unit 705 includes:

[0141] The first generating unit is used to generate a speed instruction according to a relative speed in a feature of the obstacle.

[0142] The second generating unit is used to calculate the steering angle of the vehicle according to the obstacle classification information and the position in the feature of the obstacle, and generate a steering instruction based on the steering angle.

[0143] The third generating unit is used to generate control information according to the speed instruction and the steering instruction.

[0144] It should be noted that the specific working process of each module provided in the above embodiments of the present application can refer to the corresponding steps in the above method embodiments, which will not be repeated here.

[0145] It should also be noted that the obstacle information reporting system provided in the embodiment of the present application has the technical effects of any of the above embodiments, and the embodiment of the present application will not be described in detail here.

[0146] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0147] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for reporting obstacle information, characterized in that: Applied to the FPGA transmitter, including: Get the raw data collected by multiple sensors; Preprocessing the original data to obtain target original data; Determining the characteristics of the obstacle based on the target original data; Classifying and identifying the obstacle according to the characteristics of the obstacle and the target original data to obtain classification information of the obstacle; generating control information based on the characteristics of the obstacle and the classification information, and converting the obstacle classification information and the control information into a CAN message; The CAN message is sent to a CAN controller, and the CAN controller reports the CAN message to a receiving end via a CAN bus.

2. The method according to claim 1, characterized in that The preprocessing of the original data to obtain target original data includes: Performing filtering on the original data to obtain filtered original data; The filtered original data is subjected to noise reduction processing to obtain target original data.

3. The method according to claim 1, characterized in that The step of determining the characteristics of the obstacle according to the target original data includes: Inputting the target raw data into a Kalman filter for feature prediction, and predicting the position and relative speed of the obstacle; The position and relative speed of the obstacle are determined as characteristics of the obstacle.

4. The method according to claim 1, characterized in that: The step of classifying and identifying the obstacle according to the characteristics of the obstacle and the target original data to obtain classification information of the obstacle includes: Extracting basic information of the obstacle from the target raw data; The basic information and the features of the obstacle are input into a pre-trained machine learning model, and the classification information of the obstacle is output.

5. The method according to claim 1, characterized in that The generating control information based on the feature of the obstacle and the classification information includes: generating a speed command according to a relative speed in a characteristic of the obstacle; Calculating a steering angle of the vehicle according to the obstacle classification information and the position of the feature of the obstacle, and generating a steering instruction based on the steering angle; Control information is generated according to the speed command and the steering command.

6. A reporting system for obstacle information, characterized in that: Applied to the FPGA transmitter, including: A data acquisition unit, used to acquire raw data collected by multiple sensors; A preprocessing unit, used for preprocessing the original data to obtain target original data; A feature determination unit, used to determine the features of the obstacle based on the target original data; A classification unit, configured to classify and identify the obstacle according to the characteristics of the obstacle and the target original data, and obtain classification information of the obstacle; an information generating unit, configured to generate control information based on the characteristics of the obstacle and the classification information, and convert the obstacle classification information and the control information into a CAN message; The reporting unit is used to send the CAN message to the CAN controller, and the CAN controller reports the CAN message to the receiving end through the CAN bus.

7. The system according to claim 6, characterized in that The pre-processing unit comprises: A filtering processing unit, used for filtering the original data to obtain filtered original data; The noise reduction processing unit is used to perform noise reduction processing on the filtered original data to obtain target original data.

8. The system according to claim 6, characterized in that The feature determination unit comprises: A prediction unit, used for inputting the target raw data into a Kalman filter for feature prediction, and predicting the position and relative speed of the obstacle; The feature sub-determination unit is used to determine the position and relative speed of the obstacle as the features of the obstacle.

9. The system according to claim 6, characterized in that The classification unit includes: An extraction unit, used to extract basic information of the obstacle from the target raw data; An input unit is used to input the basic information and the characteristics of the obstacle into a pre-trained machine learning model and output the classification information of the obstacle.

10. The system according to claim 6, characterized in that The information generating unit comprises: A first generating unit, configured to generate a speed instruction according to a relative speed in a characteristic of the obstacle; a second generating unit, configured to calculate a steering angle of the vehicle according to the obstacle classification information and a position in the feature of the obstacle, and to generate a steering instruction based on the steering angle; The third generating unit is used to generate control information according to the speed instruction and the steering instruction.

Images