A method and apparatus for controlling a vehicle

Abstract

The application relates to a method and device for controlling a vehicle, wherein the method comprises: generating fusion perception data according to data obtained by each perception sensor, taking time data of a preset perception sensor as perception reference time data; generating positioning data according to data obtained by each positioning sensor, taking time data of a preset positioning sensor as positioning reference time data; updating the positioning reference time data and the positioning data to obtain new map data; obtaining new map data closest to the perception reference time data as matching map data; obtaining decision planning path information according to the matching map data and the fusion perception data; or obtaining fusion perception data closest to the positioning reference time data as matching fusion perception data; obtaining decision planning path information according to the new map data and the matching fusion perception data; and performing lateral and longitudinal control on the vehicle according to positioning data closest to a current time and the decision planning path information.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and more particularly to a method for controlling a vehicle and a device for controlling a vehicle. Background Technology

[0002] Autonomous driving systems incorporate numerous perception and positioning sensors, including but not limited to LiDAR, cameras, millimeter-wave radar, GPS, inertial measurement units, and wheel speed sensors. Different sensors have different acquisition times and frequencies, necessitating time synchronization to ensure the autonomous driving system operates within a unified timeframe. This spatiotemporal synchronization is particularly crucial at high speeds, ensuring downstream decision-making and control systems determine driving trajectories simultaneously, which is vital for driving safety.

[0003] Existing technical solutions typically use the reception time of the software system as the signal time of each sensing or positioning sensor, and use this signal time as the benchmark and basis for time synchronization among the sensors. Meanwhile, in downstream decision-making and control algorithms, the transmission delay of sensor time within the system is usually not considered; instead, sensor data from a certain period ago is used as the input for current decision-making and control calculations.

[0004] However, the sensor's reception time does not represent the true time of obstacles in the environment. Taking vision as an example, a certain amount of time is required from the camera triggering the capture to image processing and information transmission, and only the camera trigger time reflects the true obstacle information. The difference between these two times leads to errors in the obstacle information processed downstream, which can affect vehicle performance in high-speed and high-yaw-rate scenarios. Furthermore, if perception processing and system transmission delays are not considered during the decision-making and control phases, it will also cause lags in decision-making and control, and even affect driving safety. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing a method and apparatus for controlling vehicles, solving the problem of overall time synchronization within an autonomous driving system. This is achieved by using a hardware controller to trigger and ensure the time consistency of each sensor, and by using software time synchronization to ensure that the vehicle makes advanced decisions and controls based on environmental information at the correct time.

[0006] To achieve the above objectives, a first aspect of the present invention provides a method for controlling a vehicle, the method comprising:

[0007] Based on the data acquired by each sensing sensor, fused sensing data is generated, and the time data in the preset sensing sensor is used as the sensing reference time data.

[0008] Positioning data is generated based on the data acquired by each positioning sensor, and the time data in the preset positioning sensor is used as the positioning reference time data.

[0009] The map data is updated based on the positioning reference time data and the positioning data to obtain new map data;

[0010] Obtain the new map data whose corresponding time data is closest to the perception reference time data from the new map data, and use it as the matching map data. Decision planning path information is then obtained based on the matching map data and the fused perception data; or

[0011] The fused sensing data that is closest to the positioning reference time data in the fused sensing data is obtained as the matching fused sensing data, and the decision planning path information is obtained based on the new map data and the matching fused sensing data.

[0012] The vehicle is controlled laterally and longitudinally based on the location data most recent to the current time and the decision-making and planning path information.

[0013] Preferably, after using the time data from the preset positioning sensor as the positioning reference time data, the method for controlling the vehicle further includes:

[0014] The positioning reference time data is sent to each processor in the vehicle via a time synchronization bus.

[0015] Preferably, the step of updating the map data based on the positioning reference time data and the positioning data to obtain new map data specifically includes:

[0016] Select the local map data to be updated from the map data according to the preset size;

[0017] Based on the positioning reference time data and the positioning data, the local map data is updated to obtain new map data.

[0018] Preferably, the newest map data most recent to the current time is selected as the first cached map data;

[0019] Select the fused sensing data that is closest to the current time as the first cached fused sensing data;

[0020] Determine whether the time data of the first cached map data is earlier than the time data of the first cached fused perception data;

[0021] When the time data of the first cached map data is later than the time data of the first cached fused sensing data, the new map data that is closest to the time data of the first cached fused sensing data is determined as the map data to be used, and the first cached fused sensing data is determined as the fused sensing data to be used.

[0022] More preferably, when the time data of the first cached map data is earlier than the time data of the first cached fused sensing data, the fused sensing data in the fused sensing data cache queue that is closest to the time data of the first cached map data is determined as the fused sensing data to be used, and the first cached map data is determined as the map data to be used.

[0023] Further preferably, the map data to be used and the fused perception data to be used are projected onto the vehicle coordinate system to obtain the decision planning path information at the first moment.

[0024] A further preferred approach is to convert the received decision-making and planning path information from the first moment into decision-making and planning path information for the current moment.

[0025] Further optimization involves acquiring the current location data and the location data from the first moment.

[0026] Calculate the difference between the current location data and the location data at the first moment, and correct the decision planning path information based on the difference.

[0027] A second aspect of the present invention provides an apparatus for controlling a vehicle, referring to any of the methods for controlling a vehicle described above, wherein the apparatus for controlling the vehicle comprises:

[0028] The sensing module is used to generate fused sensing data based on the data acquired by each sensing sensor, and uses the time data in the preset sensing sensor as the sensing reference time data.

[0029] The positioning module is used to generate positioning data based on the data acquired by each positioning sensor, and to use the time data in the preset positioning sensor as the positioning reference time data.

[0030] A map engine is used to update map data based on the positioning reference time data and the positioning data to obtain new map data;

[0031] The decision planning module is used to acquire the new map data whose corresponding time data is closest to the perception reference time data from the new map data, and use it as the matching map data; and to obtain decision planning path information based on the matching map data and the fused perception data; or

[0032] The fused sensing data that is closest to the positioning reference time data in the fused sensing data is obtained as the matching fused sensing data; the decision planning path information is obtained based on the new map data and the matching fused sensing data.

[0033] The control module is used to perform lateral and longitudinal control of the vehicle based on the location data most recent to the current time and the decision-planning path information.

[0034] Preferably, the sensing sensors include: lidar, millimeter-wave radar, and camera;

[0035] The positioning sensors include: an inertial measurement unit (IMU), a global navigation satellite system (GNSS), and a wheel speed meter.

[0036] The vehicle control method and apparatus provided in this invention improve the consistency and accuracy of fusion results through hardware controller triggering and software time synchronization, avoiding erroneous matching of multi-sensor information; improve the correctness of obstacle perception and map data matching, obtaining more accurate relative positional relationships between obstacles and lanes, thereby improving the safety and rationality of decision planning results; improve the real-time response of the autonomous driving system, improve the control lag phenomenon caused by processing delay, and obtain good vehicle control accuracy. Attached Figure Description

[0037] Figure 1 This is a first flowchart of a method for controlling a vehicle provided in an embodiment of the present invention;

[0038] Figure 2 This is a second flowchart of a method for controlling a vehicle provided in an embodiment of the present invention;

[0039] Figure 3 A time flow diagram provided for an embodiment of the present invention;

[0040] Figure 4 A logical diagram of the decision planning module provided in an embodiment of the present invention;

[0041] Figure 5 This is a third flowchart of a method for controlling a vehicle provided in an embodiment of the present invention;

[0042] Figure 6 A logical diagram illustrating the corrected decision planning information provided in this embodiment of the invention;

[0043] Figure 7 A schematic diagram of the structure of the vehicle control device provided in this embodiment of the invention. Detailed Implementation

[0044] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0045] The method and apparatus for controlling vehicles provided by this invention improve the consistency and accuracy of fusion results through hardware controller triggering and software time synchronization, avoiding erroneous matching of multi-sensor information; improve the correctness of obstacle perception and map data matching, obtain more accurate relative positional relationships between obstacles and lanes, thereby improving the safety and rationality of decision planning results; improve the real-time response of the autonomous driving system, improve the control lag phenomenon caused by processing delay, and obtain good vehicle control accuracy.

[0046] Example 1

[0047] Figure 1 This is a first flowchart of a method for controlling a vehicle provided in an embodiment of the present invention, as shown below. Figure 1 As shown, methods for controlling a vehicle include:

[0048] Step 110: Generate fused sensing data based on the data acquired by each sensing sensor, and use the time data in the preset sensing sensor as the sensing reference time data;

[0049] Specifically, the sensing sensors include, but are not limited to, LiDAR, millimeter-wave radar, and cameras. The fused sensing data includes LiDAR data, millimeter-wave data, and visual data corresponding to each sensing sensor. Based on the characteristics of each of the aforementioned sensing sensors, the sensing module selects a preset sensing sensor's time as the sensing reference time; typically, the preset sensing sensor is LiDAR.

[0050] This invention does not impose specific limitations on the number of sensing sensors, their models, or their installation locations. Furthermore, it does not impose specific limitations on the sampling frequency and data transmission method of each sensing sensor. The sensing configuration scheme of LiDAR, millimeter-wave radar, and camera in this invention is merely an example of one sensing sensor configuration scheme.

[0051] Step 120: Generate positioning data based on the data acquired by each positioning sensor, and use the time data in the preset positioning sensor as the positioning reference time data;

[0052] Specifically, positioning sensors include, but are not limited to, inertial measurement units (IMUs), global navigation satellite systems (GNSS), and wheel speed sensors. Positioning data is obtained by aggregating data from each positioning sensor. Based on the characteristics of these sensors, the positioning module selects a preset positioning sensor's time as the positioning reference time; typically, the preset positioning sensor is the GNSS.

[0053] The autonomous driving system of this invention is driven by two types of data time: perception reference time and positioning reference time. Generally, the frequency of positioning reference time is several times higher than that of perception reference time.

[0054] Step 130: Update the map data based on the positioning reference time data and positioning data to obtain new map data;

[0055] Specifically, based on a preset size, a local map data segment to be updated is selected from the map data. Then, based on the positioning reference time data and positioning data, the local map data is updated to obtain the new map data.

[0056] To ensure the timeliness of dynamic map updates, this invention selects to update local layers based on high-frequency positioning data to provide a data source for map data. Furthermore, by combining high-precision map data, the topological relationship of road structures near the location is extracted. In this way, timely and effective data can be provided to the decision-making and planning module based on the current task information.

[0057] In the preferred embodiment, after obtaining new map data, the new map data is stored. Further, the new map data can be stored in a new map data cache queue. Here, the new map data cache queue refers to the queue in which the decision-making and planning module stores each new frame of map data it receives. When the new map data queue is full, exceeding the set maximum number of frames, the oldest new map data frame in the cache queue is deleted to continue caching the latest frame. There is no specific limit to the length of the new map data cache queue. Generally, the length of the data cache queue is chosen based on the sensor's sampling frequency and the time delay between data from different sensors.

[0058] Step 140: Obtain the new map data that is closest to the perception reference time data in the new map data, and use it as the matching map data. Obtain the decision planning path information based on the matching map data and the fused perception data.

[0059] Specifically, the matching map data and the fused sensing data are the closest pair of data in time. In this invention, the matching map data and the fused sensing data are regarded as data at the same moment.

[0060] Step 150: Perform lateral and longitudinal control on the vehicle based on the location data most recent to the current time and the decision-making and planning path information.

[0061] Figure 2 This is a second flowchart of a method for controlling a vehicle provided in an embodiment of the present invention, as shown below. Figure 2 As shown, step 140 above can be replaced by step 160.

[0062] Step 160: Obtain the fusion sensing data whose corresponding time data is closest to the positioning reference time data from the fusion sensing data, and use it as the matching fusion sensing data. Obtain decision planning path information based on the new map data and the matching fusion sensing data.

[0063] Specifically, the matched sensing data and the new map data are the closest pair of data in time. In this invention, the matched sensing data and the new map data are regarded as data at the same moment.

[0064] To ensure blind-spot-free coverage of the environment surrounding autonomous vehicles, multiple LiDAR sensors are typically installed on the vehicle. For example... Figure 3 The time synchronization of the main lidar and the two blind zone lidars shown employs a trigger synchronization method for the lidar sensors. This method uses an external, identical signal source as the trigger source for sensor sampling, achieving synchronization of data from the three lidar sensors through external triggering. The final lidar sampling time data obtained by the sensing module is T. t lidar And the sampling time data T t lidar This serves as the sensing reference time data for the current moment of the sensing module.

[0065] For millimeter-wave data, the method is similar to that used for acquiring map data. The millimeter-wave data corresponding to the millimeter-wave time data that is closest to the current sensing reference time data is searched in the millimeter-wave data cache queue using the sensing module's current sensing reference time data. Figure 3 The superscript n in the time of the millimeter-wave data buffer queue represents the nth frame of millimeter-wave data acquired by the sensing module at the current moment.

[0066] For camera data, if hardware time synchronization is achieved through external triggering, and the camera's sampling frequency matches the LiDAR's sampling frequency, then software time synchronization of the camera data is not required in the perception module. Figure 3The method shown indicates that the visual time closest to the baseline time data is T. t vision If the camera is not externally synchronized, a software synchronization method similar to map data or millimeter-wave data is required.

[0067] Both map data and fused perception data surrounding the vehicle need to be synchronized in time before decision-making and planning can be performed. The goal is to obtain map data and fused perception data with the smallest time difference from the current time. The specific time synchronization logic is as follows: Figure 4 As shown, the specific process is as follows: Figure 5 As shown, the details are as follows:

[0068] Step 170: Select the newest map data closest to the current time as the first cached map data. Select the fused sensing data closest to the current time as the first cached fused sensing data.

[0069] Step 171: Determine whether the time data of the first cached map data is earlier than the time data of the first cached fused perception data.

[0070] Specifically, when the time data of the first cached map data is later than the time data of the first cached fused perception data, step 172 is executed; when the time data of the first cached map data is earlier than the time data of the first cached fused perception data, step 173 is executed.

[0071] Step 172: Determine the new map data that is closest in time to the first cached fused perception data as the map data to be used, and determine the first cached fused perception data as the fused perception data to be used.

[0072] Step 173: Determine the fusion sensing data in the fusion sensing data cache queue that is closest to the time data of the first cached map data as the fusion sensing data to be used, and determine the first cached map data as the map data to be used.

[0073] Step 174: Project the map data to be used and the fused perception data to be used onto the vehicle coordinate system to obtain the decision-making and planning path information at the first moment.

[0074] In a preferred embodiment, to prevent vehicle response lag, the present invention converts the time corresponding to the decision-planning path information and corrects the decision-planning information based on positioning data. Specifically: the received decision-planning path information at the first moment is converted into decision-planning path information at the current moment. The positioning data at the current moment and the positioning data at the first moment are obtained. The difference between the positioning data at the current moment and the positioning data at the first moment is calculated, and the decision-planning path information is corrected based on the difference.

[0075] In other words, such as Figure 4 As shown, two buffer queues are provided at the stage where the decision and planning module receives data, which respectively store map data and fused perception data within a certain time range. Each piece of data is attached with time data, wherein the time data of the fused perception data is the perception reference time data, and the time data of the map data is the corresponding positioning reference time data. Assuming that the latest time data in the buffered map data is Tm, the oldest time data is Tn, the latest time data in the buffered fused perception data is Tp, and the oldest time data is Tq, the lengths of the two buffer queues, namely the map data buffer queue and the fused perception data buffer queue, shall satisfy Tm≥Tq and Tn≤Tp.

[0076] The decision and planning module respectively retrieves the latest map data and fused perception data from the map data buffer queue and the fused perception data buffer queue, and compares their time data.

[0077] If Tm≥Tp, the map data with the time data closest to Tp is searched in the map buffer data queue as the map data to be used, and the latest fused perception data is taken as the fused perception data to be used at the same time.

[0078] If Tm<Tp, the fused perception data with the time data closest to Tm is searched in the fused perception buffer data queue as the fused perception data to be used, and the latest map data is taken as the map data to be used at the same time. The map data and fused perception data selected in the above steps are projected to the vehicle coordinate system, so as to obtain the relationship between fused perception and the map at the same moment, assumed to be under the perception reference time data Tp, and then behavior decision and motion planning are performed.

[0079] To ensure the normal operation of time synchronization for subsequent vehicle control, such as Figure 6 As shown, the decision and planning module acquires the positioning data Pp of the vehicle at the moment Tp through the location service, and transmits the positioning data Pp together with the planned path data to the control module of the vehicle.

[0080] In a preferred solution, to ensure the normal operation of time synchronization for subsequent vehicle control, after the time data in the preset positioning sensor is taken as the positioning reference time data, the positioning reference time data is transmitted to each processor of the vehicle through the time synchronization bus.

[0081] Specifically, as Figure 7 As shown, the hardware-triggered time synchronization solution is divided into processor time synchronization and sensor time synchronization, and the source of time synchronization for both is the positioning module.

[0082] The processor time synchronization positioning system uses a time synchronization bus, including but not limited to a time synchronization bus, to perform parallel synchronization on each processor. After receiving the synchronization signal, the processor immediately synchronizes its clock system, and the synchronization accuracy can reach the microsecond level.

[0083] In the preferred scheme, to ensure the stability of time synchronization, time synchronization verification is performed between each processor to guarantee the robustness of the time synchronization system.

[0084] Sensor time synchronization encompasses the time synchronization of LiDAR, millimeter-wave radar, and cameras. Different time synchronization methods are deployed based on the different characteristics of each sensor. LiDAR and millimeter-wave radar synchronize with the positioning module via a time synchronization bus. For multi-line LiDAR, its scanning characteristics dictate that its entire frame data is not generated simultaneously. When the radar rotation speed is 600 rpm, its data output frequency is 10 Hz, and the time difference between the beginning and end of a point cloud data frame is 100 ms. Through the time synchronization interface, each point cloud of the LiDAR contains scan time data, allowing the perception module to retrieve the LiDAR sensor's time data and thus align with other sensors. Millimeter-wave radar time synchronization is based on hardware characteristics, ensuring that its output fused perception includes scan time data. For camera time synchronization, since the camera requires a certain amount of time from exposure to image generation, camera time synchronization requires the processor to trigger synchronous exposure. The exposure time is configured based on the relative angle between the camera's installation position and the LiDAR, enabling the camera and LiDAR to scan the same environment simultaneously, providing the perception module with environmental information at the same moment.

[0085] Example 2

[0086] Figure 7 This is a schematic diagram of a vehicle control device provided in an embodiment of the present invention, such as... Figure 7 As shown, Embodiment 2 of the present invention provides a device for controlling a vehicle, comprising:

[0087] The sensing module 1 is used to generate fused sensing data based on the data acquired by each sensing sensor, and uses the time data from the preset sensing sensors as the sensing reference time data. The sensing sensors are preferably: LiDAR, millimeter-wave radar, and a camera.

[0088] Positioning module 2 is used to generate positioning data based on the data acquired by each positioning sensor, and to use the time data from a preset positioning sensor as the positioning reference time data. The preferred positioning sensors are: an inertial measurement unit (IMU), a global navigation satellite system (GNSS), and a wheel speedometer.

[0089] Map Engine 3 is used to update map data based on positioning reference time data and positioning data to obtain new map data.

[0090] The map engine serves as the upper application layer, including but not limited to a function library that provides map data management for the perception, positioning, and decision-making / planning modules, enabling dynamic map loading and querying. It also encapsulates the topological relationships and guidance data of high-precision map geometric data, providing route analysis, topology analysis, and guidance information to the upper layers. Through the encapsulation of operation interfaces, it provides a unified map usage interface, categorized into map services and navigation services. Map services are provided externally through interfaces, while navigation services are periodically broadcast to the decision-making / planning layer.

[0091] autonomous driving Figure 1 There are usually many layers, each used by different objects, but the requirements for the map engine are the same: to update the high-precision map as quickly and accurately as possible. Otherwise, if there is a time difference between the current information of the module and the data in the map, it will cause logical errors in the module.

[0092] Decision planning module 4 is used to acquire the new map data whose corresponding time data is closest to the perception reference time data, as the matching map data; and to obtain decision planning path information based on the matching map data and the fused perception data. Alternatively...

[0093] The fused sensing data that is closest to the positioning reference time data in the fused sensing data is obtained as the matching fused sensing data; the decision planning path information is obtained based on the new map data and the matching fused sensing data.

[0094] Control module 5 is used to perform lateral and longitudinal control of the vehicle based on the location data most recent to the current time and the decision-making and planning path information.

[0095] In a preferred embodiment, the device for controlling the vehicle further includes a multi-device time synchronization monitoring module 6, which monitors the synchronization time status of the processor and sends an alarm signal to the decision planning module 4 and / or the control module 5 when the status is abnormal, so as to ensure the robustness of time synchronization.

[0096] In a specific example, the positioning module 2 is also used to: after using the time data in the preset positioning sensor as the positioning reference time data, send the positioning reference time data to each processor of the vehicle via the time synchronization bus.

[0097] In a specific example, map engine 3 is used to: select a local map data segment from the map data to be updated based on a preset size; and update the local map data based on the positioning reference time data and positioning data to obtain new map data.

[0098] In a specific example, the decision planning module 4 is also used to: select the newest map data closest to the current time as the first cached map data; select the fusion sensing data closest to the current time as the first cached fusion sensing data; determine whether the time data of the first cached map data is earlier than the time data of the first cached fusion sensing data; when the time data of the first cached map data is later than the time data of the first cached fusion sensing data, determine the newest map data closest to the time data of the first cached fusion sensing data as the map data to be used, and determine the first cached fusion sensing data as the fusion sensing data to be used.

[0099] In a specific example, the decision planning module 4 is also used to: when the time data of the first cached map data is earlier than the time data of the first cached fused sensing data, determine the fused sensing data in the fused sensing data cache queue that is closest to the time data of the first cached map data as the fused sensing data to be used, and determine the first cached map data as the map data to be used.

[0100] In a specific example, the decision planning module 4 is also used to: project the map data to be used and the fused perception data to be used onto the vehicle coordinate system to obtain the decision planning path information at the first moment.

[0101] In a specific example, the decision planning module 4 is also used to convert the received decision planning path information at the first moment into the decision planning path information at the current moment.

[0102] In a specific example, the decision planning module 4 is also used to: acquire the current location data and the location data at the first moment; calculate the difference between the current location data and the location data at the first moment; and correct the decision planning path information based on the difference.

[0103] The method and apparatus for controlling vehicles of the present invention improve the consistency and accuracy of fusion results through hardware controller triggering and software time synchronization, avoiding erroneous matching of multi-sensor information; improve the correctness of obstacle perception and map data matching, obtain more accurate relative positional relationships between obstacles and lanes, thereby improving the safety and rationality of decision planning results; improve the real-time response of the autonomous driving system, improve the control lag phenomenon caused by processing delay, and obtain good vehicle control accuracy.

[0104] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0105] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0106] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for controlling a vehicle, characterized in that, The method for controlling the vehicle includes: Based on the data acquired by each sensing sensor, fused sensing data is generated, and the time data in the preset sensing sensor is used as the sensing reference time data. Positioning data is generated based on the data acquired by each positioning sensor, and the time data in the preset positioning sensor is used as the positioning reference time data. The map data is updated based on the positioning reference time data and the positioning data to obtain new map data; Obtain the new map data whose corresponding time data is closest to the perception reference time data from the new map data, and use it as the matching map data. Decision planning path information is then obtained based on the matching map data and the fused perception data; or The fused sensing data that is closest to the positioning reference time data in the fused sensing data is obtained as the matching fused sensing data, and the decision planning path information is obtained based on the new map data and the matching fused sensing data. The vehicle is controlled laterally and longitudinally based on the location data most recent to the current time and the decision-making and planning path information.

2. The method for controlling a vehicle according to claim 1, characterized in that, After using the time data from the preset positioning sensor as the positioning reference time data, the method for controlling the vehicle further includes: The positioning reference time data is sent to each processor in the vehicle via a time synchronization bus.

3. The method for controlling a vehicle according to claim 1, characterized in that, The step of updating map data based on the positioning reference time data and the positioning data to obtain new map data specifically includes: Select the local map data to be updated from the map data according to the preset size; Based on the positioning reference time data and the positioning data, the local map data is updated to obtain new map data.

4. The method for controlling a vehicle according to claim 1, characterized in that, The method for controlling the vehicle also includes: Select the newest map data that is closest to the current time as the first cached map data; Select the fused sensing data that is closest to the current time as the first cached fused sensing data; Determine whether the time data of the first cached map data is earlier than the time data of the first cached fused perception data; When the time data of the first cached map data is later than the time data of the first cached fused sensing data, the new map data that is closest to the time data of the first cached fused sensing data is determined as the map data to be used, and the first cached fused sensing data is determined as the fused sensing data to be used.

5. The method for controlling a vehicle according to claim 4, characterized in that, The method for controlling the vehicle also includes: When the time data of the first cached map data is earlier than the time data of the first cached fused sensing data, the fused sensing data in the fused sensing data cache queue that is closest to the time data of the first cached map data is determined as the fused sensing data to be used, and the first cached map data is determined as the map data to be used.

6. The method for controlling a vehicle according to claim 5, characterized in that, The method for controlling the vehicle also includes: The map data to be used and the fused perception data to be used are projected onto the vehicle coordinate system to obtain the decision planning path information at the first moment.

7. The method for controlling a vehicle according to claim 6, characterized in that, The method for controlling the vehicle also includes: The received decision-making and planning path information from the first moment is converted into decision-making and planning path information for the current moment.

8. The method for controlling a vehicle according to claim 7, characterized in that, The method for controlling the vehicle also includes: Obtain the current location data and the location data from the previous moment; Calculate the difference between the current location data and the location data at the first moment, and correct the decision planning path information based on the difference.

9. An apparatus for controlling a vehicle, comprising the method for controlling a vehicle according to any one of claims 1-8, characterized in that, The device for controlling the vehicle includes: The sensing module is used to generate fused sensing data based on the data acquired by each sensing sensor, and uses the time data in the preset sensing sensor as the sensing reference time data. The positioning module is used to generate positioning data based on the data acquired by each positioning sensor, and to use the time data in the preset positioning sensor as the positioning reference time data. A map engine is used to update map data based on the positioning reference time data and the positioning data to obtain new map data; The decision planning module is used to acquire the new map data whose corresponding time data is closest to the perception reference time data from the new map data, and use it as the matching map data; and to obtain decision planning path information based on the matching map data and the fused perception data; or The fused sensing data that is closest to the positioning reference time data in the fused sensing data is obtained as the matching fused sensing data; the decision planning path information is obtained based on the new map data and the matching fused sensing data. The control module is used to perform lateral and longitudinal control of the vehicle based on the location data most recent to the current time and the decision-planning path information.

10. The device for controlling a vehicle according to claim 9, characterized in that, The sensing sensors include: lidar, millimeter-wave radar, and camera; The positioning sensors include: an inertial measurement unit (IMU), a global navigation satellite system (GNSS), and a wheel speed meter.

Images