Intelligent driving method and apparatus

Abstract

An intelligent driving method and apparatus, an intelligent driving system, a computer-readable storage medium, a computer program product, a chip, and a vehicle. The intelligent driving method comprises: acquiring surrounding environment information of a vehicle; on the basis of the surrounding environment information, planning a trajectory along which the vehicle returns to a main road from the pose of the vehicle, wherein the pose of the vehicle is outside the main road, and the main road comprises a main road which supports an intelligent driving function and enables the vehicle to travel to a navigation end point; and controlling the vehicle to return to the main road along the planned trajectory. The method can be applied to a new energy vehicle or an intelligent driving vehicle. By means of planning the trajectory along which the vehicle returns to the main road and controlling the vehicle to return to the main road, the vehicle can activate the intelligent driving function in any pose, so that usage scenarios of the intelligent driving function can be expanded, and the usage experience of a user can be improved.

Description

Methods and devices for intelligent driving

[0001] This application claims priority to Chinese Patent Application No. 202410515189.1, filed on April 25, 2024, entitled "Method and Apparatus for Intelligent Driving", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of intelligent driving, and more specifically, to a method and apparatus for intelligent driving. Background Technology

[0003] With technological advancements, intelligent driving functions are becoming increasingly widespread. Currently, activating certain intelligent driving functions requires meeting specific conditions. These include whether the vehicle's doors, trunk lid, and other opening and closing mechanisms are closed, and whether the vehicle is on a main road. When the vehicle's position does not meet these activation conditions, manual driving by the user is required until the conditions are met, significantly impacting the user experience.

[0004] Therefore, how to activate the intelligent driving function when the vehicle's position does not meet the activation conditions becomes a problem that needs to be solved. Summary of the Invention

[0005] This application provides a method and apparatus for intelligent driving, which enables a vehicle to activate intelligent driving functions in any position, expands the application scenarios of intelligent driving functions, and improves the user experience.

[0006] Firstly, a method for intelligent driving is provided. This method can be executed by a computing platform (such as an autonomous driving domain controller), or by a processor, processing circuitry, or chip within the computing platform, or by an intelligent driving system or vehicle equipped with the computing platform.

[0007] The method includes: acquiring information about the vehicle's surrounding environment; planning a first trajectory for the vehicle to return to the main road from a first pose based on the surrounding environment information, wherein the first pose is outside the main road and the main road includes roads that support intelligent driving functions and can be driven to the navigation destination; and controlling the vehicle to return to the main road along the first trajectory.

[0008] In real-world scenarios, vehicles may be in any position. However, for some intelligent driving functions, the vehicle's position must meet certain conditions (e.g., the vehicle must be on a main road) to be activated. In this application, when the vehicle is not on a main road, a first trajectory to return to the main road is planned based on surrounding environmental information, and the vehicle is controlled to return to the main road along the first trajectory. This avoids the situation where the user has to manually drive the vehicle to a main road to activate the intelligent driving function, expands the application scenarios of the intelligent driving function, and improves the user experience.

[0009] In conjunction with the first aspect, in some implementations of the first aspect, planning the first trajectory of the vehicle returning from the first pose to the main road based on the surrounding environment information may include: determining the connection points on the main road based on the surrounding environment information; and planning the first trajectory of the vehicle returning from the first pose to the main road based on the connection points.

[0010] By determining the connection points, the process of the vehicle traveling from the first position to the navigation endpoint can be divided into two parts: the process of the vehicle traveling along the first trajectory to the main road, and the process of cruising along the main road towards the navigation endpoint. Setting reasonable connection points facilitates the separate control of these two processes and the connection / switching between them.

[0011] In conjunction with the first aspect, in some implementations of the first aspect, the first trajectory includes a connecting trajectory ending at a connecting point on the main road, in which the vehicle travels using either forward or reverse gear. The method may further include: while the vehicle is traveling along the connecting trajectory, controlling the vehicle to switch to traveling along the main road when a switching condition is met. The switching condition includes at least one of the following: the second pose of the vehicle meets preset conditions; the lateral distance between the second pose of the vehicle and the centerline of the main road is less than or equal to a first threshold; and the angle between the longitudinal axis of the vehicle and the first travel direction of the main road is less than or equal to a second threshold.

[0012] In this application, when the vehicle is traveling along the connecting trajectory, the trajectory is switched when the switching conditions are met, so that the vehicle can achieve the switching process with a small steering angle, which can achieve a smooth transition and reduce the impact of the vehicle's steering on the ride comfort during the switching process. This method can achieve seamless switching and improve the ride experience of users in the cabin.

[0013] In conjunction with the first aspect, in some implementations of the first aspect, the switching condition is met at the first moment within the first time period. The method may also include: controlling the vehicle speed to change continuously within the first time period.

[0014] In this application, controlling the vehicle to change continuously for a period of time before and after the switch can reduce the impact of sudden speed changes on the user's riding experience and improve the riding comfort of the users in the cabin.

[0015] In conjunction with the first aspect, in some implementations of the first aspect, the vehicle's acceleration changes continuously at the first moment.

[0016] When acceleration changes continuously, the vehicle's acceleration will not experience sudden changes or abrupt jumps, and the vehicle's speed will change smoothly and continuously. This avoids the jerking sensation and discomfort caused by sudden acceleration changes, achieving a seamless transition and improving passenger comfort.

[0017] In conjunction with the first aspect, in some implementations of the first aspect, the first waypoint is the endpoint of the shortest travel trajectory from the first pose back to the main road when there are no surrounding obstacles. Determining the connection point on the main road may include: translating the first waypoint along the extension direction of the main road and / or a direction orthogonal to the extension direction to determine candidate waypoints with reachable trajectories, where the vehicle has no risk of collision with surrounding obstacles when returning to the main road from the first pose along the reachable trajectory; and determining the connection point based on the candidate waypoints with reachable trajectories.

[0018] In conjunction with the first aspect, in some implementations of the first aspect, candidate waypoints with reachable trajectories may include a first candidate waypoint. Determining a connecting point based on candidate waypoints with reachable trajectories may include: planning a first candidate trajectory from the first position to the first candidate waypoint; and determining whether to use the first candidate waypoint as a connecting point based on the cost of the first candidate trajectory. The cost of the first candidate trajectory may be determined based on at least one of the length of the first candidate trajectory, the number of gear shifts during the vehicle's journey along the first candidate trajectory, and the number of steering direction changes during the vehicle's journey along the first candidate trajectory.

[0019] In manual driving scenarios, there might be a preference for using fewer gear shifts to navigate the vehicle to the main road and for employing fewer S-shaped trajectories. In this application, by considering the cost of candidate trajectories and the number of gear shifts and steering direction changes corresponding to each trajectory, the human-likeness of the first trajectory can be improved. This enhances the passenger comfort in the cabin, helps drivers of other vehicles anticipate the vehicle's driving behavior, and improves driving safety.

[0020] In conjunction with the first aspect, in some implementations of the first aspect, the second waypoint is the waypoint on the main road that is closest to the first pose, and the connection point is located within the first range corresponding to the second pose.

[0021] In conjunction with the first aspect, in some implementations of the first aspect, the method may further include: determining a first vacant area based on surrounding environmental information, the first vacant area being used to plan a first candidate trajectory, the first vacant area including at least one of the following: vacant parking space, vacant intersection, and wide lane.

[0022] In conjunction with the first aspect, in some implementations of the first aspect, the method may further include: determining that the first pose does not meet preset conditions. The preset conditions may include: that there is no risk of collision with surrounding obstacles when the vehicle returns to the main road from the first pose along a trajectory planned in a preset manner.

[0023] In conjunction with the first aspect, in some implementations of the first aspect, the first direction of travel on the main road is from the first waypoint on the main road closest to the vehicle to the navigation destination.

[0024] Secondly, an intelligent driving device is provided, which can be a computing platform, or a processor, processing circuit, unit or chip in the computing platform.

[0025] The device includes an acquisition unit and a processing unit. The acquisition unit is used to acquire information about the vehicle's surrounding environment. The processing unit is used to: based on the surrounding environment information, plan a first trajectory for the vehicle to return to the main road from a first position, where the first position is outside the main road, and the main road includes roads that support intelligent driving functions and can lead to the navigation destination; and control the vehicle to return to the main road along the first trajectory.

[0026] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit can be used to: determine the connection points on the main road based on the surrounding environmental information; and plan the first trajectory based on the connection points.

[0027] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit can also be used to: control the vehicle to travel along the main road while the vehicle is traveling along the connecting trajectory, when the switching conditions are met. The switching conditions include at least one of the following: the second pose of the vehicle meets preset conditions; the lateral distance between the second pose of the vehicle and the centerline of the main road is less than or equal to a first threshold; and the angle between the longitudinal axis of the vehicle and the first travel direction of the main road is less than or equal to a second threshold.

[0028] In conjunction with the second aspect, in some implementations of the second aspect, the switching condition is met at the first moment within the first time period. The processing unit can also be used to: control the vehicle speed to change continuously within the first time period.

[0029] In conjunction with the second aspect, in some implementations of the second aspect, the vehicle's acceleration changes continuously at the first moment.

[0030] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit may be used to: translate the first waypoint along the extension direction of the main road and / or the orthogonal direction of the extension direction to determine candidate waypoints with reachable trajectories, such that the vehicle has no risk of collision with surrounding obstacles when returning to the main road from the first pose along the reachable trajectory; and determine the connection point based on the candidate waypoints with reachable trajectories.

[0031] In conjunction with the second aspect, in some implementations of the second aspect, candidate waypoints with reachable trajectories may include a first candidate waypoint. The processing unit can be used to: plan a first candidate trajectory from the first pose to the first candidate waypoint; and determine, based on the cost of the first candidate trajectory, whether to use the first candidate waypoint as a connection point.

[0032] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit can also be used to: determine a first vacant area based on surrounding environmental information, the first vacant area being used to plan a first candidate trajectory, the first vacant area including at least one of the following: vacant parking space, vacant intersection, and wide lane.

[0033] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit can also be used to: determine that the first pose does not meet preset conditions. The first pose meeting preset conditions may include: when the vehicle returns to the main road from the first pose along a trajectory planned in a preset manner, there is no risk of collision with surrounding obstacles.

[0034] Thirdly, an apparatus is provided, comprising: a memory for storing a computer program; and a processor for executing the computer program stored in the memory, such that the apparatus performs the methods described in the first aspect and any possible implementation thereof.

[0035] Fourthly, a computer program product is provided, comprising: computer program code, which, when executed on a computer, causes the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0036] Fifthly, a computer-readable storage medium is provided, which stores a computer program that, when run on a computer, causes the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0037] In a sixth aspect, a chip is provided, the chip including circuitry for performing the methods described in the first aspect and any possible implementation thereof.

[0038] In a seventh aspect, an intelligent driving system is provided, which includes the means described in the second or third aspect above and any possible implementation thereof.

[0039] Eighthly, an intelligent driving device is provided, which includes the means of the second or third aspect and any possible implementation thereof, or includes the system of the seventh aspect and any possible implementation thereof.

[0040] In one embodiment, the intelligent driving device can be a vehicle. Attached Figure Description

[0041] Figure 1 is a functional block diagram of an intelligent driving device provided in an embodiment of this application;

[0042] Figure 2 is a flowchart illustrating an intelligent driving method provided in an embodiment of this application;

[0043] Figure 3 is a schematic diagram of an intelligent driving scenario provided in an embodiment of this application;

[0044] Figure 4 is a flowchart illustrating a control method provided in an embodiment of this application;

[0045] Figure 5 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0046] Figure 6 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0047] Figure 7 is a schematic diagram of the positions of several vehicles provided in the embodiments of this application;

[0048] Figure 8 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0049] Figure 9 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0050] Figure 10 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0051] Figure 11 is a schematic diagram of another intelligent driving scenario provided in an embodiment of this application;

[0052] Figure 12 is a schematic block diagram of an apparatus provided in an embodiment of this application;

[0053] Figure 13 is a schematic block diagram of another device provided in an embodiment of this application. Detailed Implementation

[0054] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0055] Figure 1 is a functional block diagram of an intelligent driving device 100 provided in an embodiment of this application. The intelligent driving device 100 may include a perception system 120 and a computing platform 150. The perception system 120 may include one or more sensors for sensing information about the environment surrounding the intelligent driving device 100. For example, the perception system 120 may include a positioning system, which may be a Global Positioning System (GPS), a BeiDou Navigation Satellite System, or other positioning systems. The perception system 120 may also include one or more of the following: an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0056] Some or all of the functions of the intelligent driving device 100 can be controlled by the computing platform 150. The computing platform 150 may include one or more processors, such as processors 151 to 15n (n being a positive integer). A processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a type of microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field-programmable gate array (FPGA). In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement some or all of the functions of the aforementioned units. In addition, it can also be hardware circuitry designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), tensor processing unit (TPU), deep learning processing unit (DPU), etc. Furthermore, the computing platform 150 may also include a memory for storing instructions, and some or all of the processors 151 to 15n can call the instructions in the memory to implement the corresponding functions.

[0057] With the development of intelligent driving technology, vehicles are gradually evolving from purely manual driving to autonomous driving. Current and future vehicles may include one or more levels of autonomous driving, L0-L5, which are based on the classification standards of the Society of Automotive Engineers (SAE). L0 is no automation; L1 is driver assistance; L2 is partial automation; L3 is conditional automation; L4 is high automation; and L5 is full automation. Levels L1 to L3 involve monitoring road conditions and reacting to them, which can be accomplished jointly by the driver and the system, but require the driver to take over dynamic driving tasks. Levels L4 and L5 allow the driver to completely transition into a passenger role. For example, through a computing platform 150 (or a portion of the processors within the computing platform 150) and a perception system 120, an intelligent driving device 100 can achieve corresponding levels of intelligent driving functions, such as L2 to L5 autonomous driving functions. Alternatively, the intelligent driving device 100 can be a vehicle.

[0058] Currently, certain activation conditions are often required to activate some intelligent driving functions. For example, activation conditions may include whether the vehicle's doors, trunk lid, or other opening and closing components are closed. Among these, the vehicle's position and posture (referred to as pose) are crucial in determining whether the intelligent driving function can be activated (e.g., whether the vehicle is on a road that supports intelligent driving). However, in real-world scenarios, drivers may drive the vehicle to areas outside of roads supporting intelligent driving functions, or even outside the coverage area of ​​the map used by the intelligent driving system. This can lead to situations where the vehicle's position and / or posture do not meet the activation conditions when the user intends to activate certain intelligent driving functions. For example, taking the automated valet parking (AVP) function as an example, if the user intends to use the AVP function to park or intelligently recall the vehicle, and the vehicle is not on a road supporting intelligent driving or even outside the coverage area of ​​the map used by the system, the vehicle's pose will not meet the activation conditions, preventing the function from being activated. In the traditional way, the driver needs to drive the vehicle to a road that supports intelligent driving functions so that the vehicle's position meets the activation conditions before the intelligent driving function can be activated.

[0059] In view of this, embodiments of this application provide a method and apparatus for intelligent driving, which enables the intelligent driving function to be activated when the vehicle is in any position, expands the application scenarios of the intelligent driving function, improves the user experience, and is applicable to vehicles with L2 to L5 level autonomous driving functions.

[0060] For example, Figure 2 is a flowchart illustrating a method for intelligent driving provided in an embodiment of this application. This method 200 can be executed by a control device (e.g., a computing platform 150), or by a unit or module within the control device, or by a processor (e.g., processor 15n), processing circuit, or chip within the control device, or by a system or vehicle equipped with the control device. The method 200 may include the following steps:

[0061] S210, obtains information about the vehicle's surrounding environment.

[0062] Taking the intelligent driving device 100 as an example, the surrounding environment information can be collected by the perception system 120.

[0063] For example, based on the data collected by the perception system 120, it can be determined that the vehicle's initial pose is outside the main road. For instance, data collected by sensors such as cameras and radar can determine that the vehicle's pose is outside the main road. Alternatively, data from IMU and GPS can be used to determine the vehicle's pose, and combined with the map used by the vehicle, it can be determined that the vehicle's pose is outside the main road. The map used by the vehicle can include any of the following: a standard definition map (SD map), a high definition map (HD map), or a self-constructed map.

[0064] In some embodiments, multiple main roads may exist, and a suitable main road can be selected based on factors such as road congestion, expected travel time, waypoints, and user instructions. A main road can represent a road that supports intelligent driving functions and can lead to the navigation destination. The first driving direction of the main road can be from the road point on the main road closest to the vehicle / first position towards the navigation destination. The main road and its first driving direction are briefly explained below with reference to Figure 3.

[0065] For example, Figure 3 is a schematic diagram of an intelligent driving scenario provided by an embodiment of this application. In the scenario shown in Figure 3, vehicle A is outside the road that supports intelligent driving functions. It is assumed that in the scenario shown in Figure 3, the vehicle travels on the roads in a right-hand traffic manner. Figure 3(a) shows the road that supports intelligent driving functions in this scenario, and Figures 3(b) and (c) show the trajectory of vehicle A returning from pose A to different main roads in this scenario. Figure 3(a) shows the extension direction of each road, but the width of each road is not shown; Figure 3(d), taking the road where waypoint 2-2 is located as an example, shows the extension direction and width direction of the road.

[0066] In one embodiment, in the scenario shown in Figure 3, vehicle A can travel to the navigation destination along different roads in multiple ways. That is, from the multiple roads shown in Figure 3(a), one or more main roads (e.g., main road 1, main road 2) that can lead to the navigation destination can be determined. For example, waypoint 1-1 and waypoint 2-1 can be the waypoints on their respective roads that are closest to vehicle A's current pose A. Alternatively, waypoint 1-1 and waypoint 2-1 can be considered as the starting waypoints of main road 1 and main road 2, respectively. Furthermore, the position where vehicle A returns to the road containing waypoint 2-1 (e.g., waypoint 2-3) can be considered as the starting waypoint of main road 2.

[0067] In another embodiment, when main road 2 is more congested than main road 1, vehicles can intelligently select main road 1 to navigate to the navigation destination.

[0068] For example, driving direction 1 and driving direction 2 can correspond to the first driving direction of main road 1 and main road 2, respectively. Taking main road 1 as an example, the specific manifestation of driving direction 1 may differ at different road points on main road 1. For example, at road point 1-1, driving direction 1 may be manifested as driving from road point 1-1 upwards. As another example, at other road points on main road 1, driving direction 1 may be manifested as driving from that road point to its left.

[0069] S220 plans the first trajectory for the vehicle to return to the main road from its first position, based on information about the vehicle's surrounding environment.

[0070] In some embodiments, planning a first trajectory from the first pose back to the main road based on surrounding environmental information may include: determining connection points on the main road based on surrounding environmental information; and planning a first trajectory from the first pose back to the main road based on the connection points.

[0071] For example, one or more reachable trajectories may exist between the first pose and the connection point, enabling the vehicle to travel from the first pose to the connection point in an intelligent driving manner. The reachable trajectory between two positions / poses can represent that the vehicle can travel along the trajectory from one position / pose to another without risk of collision with surrounding obstacles during the journey.

[0072] In one embodiment, as shown in Figure 3(c), based on the positions of obstacles in the surrounding environment, a waypoint (such as waypoint 2-3) located on the road where waypoint 2-1 is located and to the left of waypoint 2-1 can be identified as a connection point. There can be one or more reachable trajectories between pose A and waypoint 2-3 (only one is shown in Figure 3(c)), and the vehicle can travel from pose A to the connection point along any reachable trajectory.

[0073] For example, the shortest travel trajectory for the vehicle to return to the main road from the first pose can be determined while ignoring surrounding obstacles. The endpoint of this trajectory is denoted as the first waypoint, and connection points can be determined based on this first waypoint. For instance, the first waypoint can be translated along the extension direction of the main road and / or a direction orthogonal to the extension direction of the main road to determine candidate waypoints with reachable trajectories; connection points can be determined based on the candidate waypoints with reachable trajectories. The orthogonal direction of the extension direction of the main road can also be referred to as the orthogonal direction of the main road. A candidate waypoint with a reachable trajectory indicates that a reachable trajectory exists between the candidate waypoint and the first pose.

[0074] For example, ignoring obstacles 1 and 2, the shortest travel trajectory of vehicle A from pose A to main road 2 can be determined; assuming the endpoint of this trajectory is waypoint 2-2, waypoint 2-2 can be used as an example of a first waypoint. As another example, as shown in Figures 3(c) and (d), on the road where waypoint 2-2 is located, main road 2 extends to the left and right (i.e., along directions 1 and 2); as shown in Figure 3(d), the orthogonal directions of this extension are the up and down directions (i.e., directions 3 and 4). As another example, as shown in Figure 3(d), by translating waypoint 2-2 along directions 1 to 4 respectively, candidate waypoints 1 to 4 can be obtained. As yet another example, considering obstacles 1 and 2, by determining whether there is a reachable trajectory between pose A and candidate waypoints 1 to 4, a candidate waypoint among candidate waypoints 1 to 4 with a reachable trajectory can be used as a connecting point. For example, when there are multiple candidate waypoints with reachable trajectories among candidate waypoints 1 to 4, one of them can be randomly selected as the connecting point. For another example, for these multiple candidate waypoints with reachable trajectories, by determining the cost of the trajectory from pose A back to the main road via each candidate waypoint, the candidate waypoint corresponding to the trajectory with the lowest cost can be selected as the connecting point.

[0075] In one embodiment, determining the trajectory and cost corresponding to each candidate waypoint requires a significant amount of computation. To reduce computation and improve processing speed, waypoint 2-2 can be translated along a certain direction in a priority order (e.g., translating waypoint 2-2 sequentially from direction 1 to direction 4 to obtain candidate waypoints) to obtain candidate waypoints in that direction, and it can be determined whether the obtained candidate waypoints have reachable trajectories. If a candidate waypoint in that direction does not have a reachable trajectory, then candidate waypoints in the next direction are searched. For example, waypoint 2-2 can be translated along direction 1 to obtain candidate waypoint 1. If candidate waypoint 1 has a reachable trajectory, it can be determined as a connecting point. As shown in Figure 3(c), waypoint 2-3, located in direction 1 of waypoint 2-2, can be used as an example of candidate waypoint 1. Waypoint 2-3 has a reachable trajectory with pose A. After translating waypoint 2-2 along direction 1 to obtain this candidate waypoint with a reachable trajectory, this candidate waypoint can be directly used as the connection point. There is no need to continue translating waypoint 2-2 along directions 2, 3, and 4 in priority order to find other candidate waypoints, thus reducing computational load. For example, if pose A and candidate waypoint 1 do not have a reachable trajectory, waypoint 2-2 can be translated along direction 2 in priority order to obtain candidate waypoint 2. When candidate waypoint 2 has a corresponding reachable trajectory, it can be determined as the connection point. For example, if candidate waypoint 2 also does not have a corresponding reachable trajectory, waypoint 2-2 can be translated along directions 3 and 4 to obtain the corresponding candidate waypoints and determine the connection point.

[0076] In this embodiment, when determining the connection point, based on a set priority order, the first waypoint can be translated first along the extension direction of the main road (e.g., direction 1, direction 2), and then translated along the orthogonal direction of the main road (e.g., direction 3, direction 4). Since different directions have different priorities, during the process of translating the first waypoint according to the priority order to find candidate waypoints with reachable trajectories, the first candidate waypoint with a reachable trajectory can be used as the connection point, and the search for other candidate waypoints can be stopped. Furthermore, the cost of the reachable trajectory can be determined. If its cost is greater than or equal to a cost threshold, the first waypoint can continue to be translated according to the above priority order to find other candidate waypoints with reachable trajectories whose costs meet the screening requirements.

[0077] In another embodiment, the road where the first waypoint is located can be curved, and the road can have different orthogonal directions at different waypoints on the road.

[0078] In another embodiment, in the scenario of Figure 3, assuming waypoint 2-3 is determined as the connection point, a first trajectory can be planned to return from pose A to waypoint 2-3, and a navigation path on main road 2 between waypoint 2-3 and the navigation endpoint can be planned. For example, in this first trajectory, the vehicle can make one or more turns. Also, for example, in part of the first trajectory, the vehicle may be traveling in drive gear; in part of the first trajectory, the vehicle may be traveling in reverse gear, as shown in Figure 3(c).

[0079] In another embodiment, candidate road points can be obtained by translating the road point closest to the vehicle on the main road (e.g., road point 2-1) along the extension direction of the main road (e.g., direction 1, direction 2) and / or orthogonal direction (e.g., direction 3, direction 4).

[0080] For example, candidate waypoints may include a first candidate waypoint. The first candidate waypoint can be any candidate waypoint. Determining the connection point based on the candidate waypoints may include: planning a first candidate trajectory from the first pose back to the main road based on the first candidate waypoint; and determining whether to use the first candidate waypoint as the connection point based on the cost of the first candidate trajectory. The vehicle can return to the main road from the first pose via the first candidate waypoint along the first candidate trajectory; when the vehicle returns to the main road via the first candidate trajectory, there is no risk of collision with surrounding obstacles.

[0081] The cost of a trajectory can be determined based on at least one of the following: the length of the trajectory, the number of gear shifts the vehicle makes while traveling along the trajectory, and the number of steering direction changes the vehicle makes while traveling along the trajectory.

[0082] In one embodiment, given the same trajectory length, the more gear shifts (e.g., the number of shifts between forward and reverse gears) a trajectory has, the higher the cost of that trajectory.

[0083] In another embodiment, the change of steering direction may include: the vehicle changing from turning left to turning right, or the vehicle changing from turning right to turning left. For example, when a certain trajectory is an S-shaped curve, the vehicle involves at least one change of steering direction while traveling along that trajectory.

[0084] In manual driving scenarios, there is a greater tendency to use fewer gear shifts to drive the vehicle onto the main road and to employ fewer S-shaped trajectories. In this embodiment, by considering the cost of candidate trajectories and the number of gear shifts and steering direction changes corresponding to the trajectory, the human-likeness of the determined first trajectory can be improved. This can enhance the ride comfort of the user in the cabin, help other drivers predict the vehicle's movement status, and improve driving safety.

[0085] In some possible implementations, a first vacant area can be determined based on the surrounding environment. This first vacant area can be used to plan a first candidate trajectory. For example, the first vacant area may include vacant parking spaces, vacant intersections, wide lanes, etc.

[0086] For example, the waypoint closest to the vehicle's current position on the main road can be denoted as the second waypoint. Connecting points can be located within the first range of the second waypoint. For instance, for main road 2, waypoint 2-1 can be considered an example of a second waypoint. Furthermore, the first range of waypoint 2-1 can include areas where the distance to waypoint 2-1 is less than 15 meters (or other values, such as 12 meters or 18 meters). For another example, assuming candidate waypoints 1 to 4 all have reachable trajectories, if candidate waypoint 4 is outside the first range of waypoint 2-1, while candidate waypoint 1-3 is within the first range of waypoint 2-1, connecting points can be determined from candidate waypoint 1-3. This method avoids excessive computational load and time spent searching for candidate waypoints in a particular direction, improving the efficiency of determining connecting points.

[0087] In some embodiments, searching for candidate waypoints with reachable trajectories and determining connection points may require significant computation. To reduce computation and improve efficiency, when the vehicle's pose meets preset conditions, a trajectory returning to the main road from that pose can be planned in a preset manner; while when the vehicle's pose does not meet the preset conditions, the first trajectory can be planned based on the connection points.

[0088] For example, for a certain pose, for a trajectory planned in a preset manner (e.g., by using straight lines, arcs, or arcs / line segments to connect), when the vehicle returns to the main road from the pose along the trajectory without any risk of collision with surrounding obstacles, the pose can be considered to meet the preset conditions.

[0089] In one embodiment, as shown in Figure 3(b), the trajectory of the vehicle returning from pose A to the road where waypoint 1-1 is located can be planned in an arc manner, where the steering wheel angle corresponding to the arc is less than or equal to the maximum steering wheel angle. During the journey along this trajectory, if there is no risk of collision between vehicle A and surrounding obstacles (such as obstacle 1), the pose A can be considered to meet the preset conditions.

[0090] In another embodiment, in the scenario shown in Figure 3, the trajectory of the vehicle returning from pose A to the road where waypoint 1-1 is located can also be planned by splicing arcs and line segments, or by splicing arcs with arcs. When planning this trajectory, the splicing method between arcs and arcs / line segments can be preset; the radius / arc length of the arcs and the length of the line segments used in the trajectory can be determined based on the positional relationship between pose A and main road 1.

[0091] For example, for a certain pose, if there is a risk of collision with surrounding obstacles during the vehicle's return to the main road along the trajectory planned in a preset manner, the pose can be considered not to meet the preset conditions. For example, in the scenario shown in Figure 3, for the main road 2, when vehicle A returns to the road where waypoint 2-1 is located along a trajectory planned in a preset manner (e.g., a straight line, an arc, or a combination of arcs and arcs / line segments) (not shown in Figure 3), if there is a risk of collision with obstacle 1, the pose A can be considered not to meet the preset conditions. In conventional solutions, when the vehicle's pose does not meet the preset conditions, the activation conditions for certain intelligent driving functions cannot be met; to activate the intelligent driving function, the user needs to drive the vehicle from that pose to the main road. However, in this embodiment, the vehicle can return to the main road via a first trajectory in an intelligent driving manner.

[0092] S230, control the vehicle to return to the main road along the first trajectory.

[0093] By planning a first trajectory, the vehicle can be controlled to return to the main road along that trajectory in an intelligent driving manner.

[0094] For example, the first trajectory may include a connecting trajectory ending at a connecting point on the main road. The vehicle can travel in either forward or reverse gear along the connecting trajectory. That is, the vehicle uses either forward or reverse gear along the connecting trajectory and is able to complete the journey in that gear.

[0095] In one embodiment, taking the scenario shown in Figure 3 as an example, and using waypoint 2-3 as the connecting point, the explanation is as follows. Assume that the first trajectory between pose A and waypoint 2-3 includes three segments, as shown in Figure 3(c). In the first segment, vehicle A can move from pose A in forward gear to a certain pose; in the second segment, vehicle A can move from that pose in reverse gear to another pose; in the third segment, vehicle A can move in forward gear to waypoint 2-3. This third segment can correspond to the connecting trajectory.

[0096] For example, while the vehicle is traveling along the connecting trajectory, when a switching condition is met, the vehicle can be controlled to switch from the connecting trajectory to traveling along the main road. This switching condition may include at least one of the following: the vehicle's second position meets a preset condition; the lateral distance between the vehicle's second position and the centerline of the main road is less than or equal to a first threshold; and the angle between the vehicle's longitudinal axis and the first travel direction of the main road is less than or equal to a second threshold. For example, the first threshold may be 0.3 meters or 0.5 meters. Another example is that the second threshold may be 5 degrees or 10 degrees. Furthermore, the above thresholds can be set according to actual needs.

[0097] In one embodiment, while the vehicle is traveling along the connecting trajectory, when the second pose of the vehicle meets preset conditions, the vehicle can return to the main road from the second pose using a trajectory planned in a preset manner (e.g., using a straight line, an arc, or a combination of arcs and arcs / line segments). By switching trajectories, the vehicle can return to the main road along the trajectory planned in the preset manner, and then travel along the main road.

[0098] The process of a vehicle traveling from its initial position to the navigation endpoint includes at least the process of the vehicle returning to the main road along a first trajectory from the initial position, and the process of traveling towards the navigation endpoint along the main road. The switching / connection of the vehicle's operating state between these two parts of the movement process will affect the user's riding experience. In this embodiment, during the vehicle's travel along the connecting trajectory, the trajectory is switched when the above-mentioned switching conditions are met, allowing the vehicle to achieve the switching process with a small steering angle, achieving a smooth transition, reducing the impact of the vehicle's steering on riding comfort during the switching process. This method can achieve a seamless switching and improve the riding experience of the users in the cabin.

[0099] In some embodiments, the vehicle meets the switching condition at a first moment, which falls within a first time period (e.g., a duration of 5 seconds, 10 seconds, or the time period from the start of driving in pose A until switching to the main road). The method may further include controlling the vehicle speed to change continuously within the first time period. Continuous change within a time period means that there are no abrupt or sudden changes during that period. For example, continuous change in vehicle speed can include: the vehicle speed continuously changing between different values, or the vehicle speed remaining at a certain value.

[0100] In this embodiment of the application, by controlling the vehicle speed to change continuously within a certain period before and after the switch, the impact of sudden speed changes on the user's riding experience can be reduced, which is conducive to improving the riding comfort of users in the cabin.

[0101] In some embodiments, the vehicle's jerk changes continuously at the first moment. For example, during the first time period, the vehicle's jerk may change continuously between different values, or it may remain at a certain value.

[0102] In this embodiment, the acceleration of the vehicle changes continuously before and after the switch, which can avoid the discomfort caused by sudden changes in acceleration and achieve a seamless switch, thereby improving the riding comfort of the users in the cabin.

[0103] In some embodiments, the vehicle travels from a third pose to a first pose along a planned path. The planned path may include at least one of a portion of the navigation path on a main road, a second trajectory, and a third trajectory.

[0104] The second trajectory may include: when the third pose meets the preset conditions, a trajectory planned in a preset manner (e.g., a straight line, an arc, or a combination of arcs and arcs / line segments) to return from the third pose to the main path.

[0105] The third trajectory may include: when the third pose does not meet preset conditions, a trajectory planned based on surrounding environmental information to return from the third pose to the main road. For example, determining corresponding connection points based on surrounding obstacles; and planning the vehicle's trajectory to return from the third pose to the main road based on these connection points.

[0106] For example, the portion of the navigation path on the main road prior to the navigation endpoint can include: a portion of the navigation path on the main road between the endpoint of the second trajectory and the navigation endpoint, or a portion of the navigation path on the main road between the endpoint of the third trajectory and the navigation endpoint. For instance, as shown in Figure 3(c), assuming waypoints 2-3 are connecting points, during the process of vehicle A returning to main road 2 along the trajectory shown in Figure 3(c), on the one hand, when the pose of vehicle A changes, the area that its perception system can detect changes accordingly, and the obstacles within the system's perception range are updated accordingly; on the other hand, the positions of some obstacles (such as obstacle 2) may change. In these situations, the vehicle may be unable to return to the main road along the originally planned trajectory; the connecting point can be re-determined based on the updated surrounding obstacle information, and a trajectory to return to the main road can be planned. In this scenario, the portion of the vehicle that has already traveled along the originally planned trajectory can correspond to the third trajectory; the trajectory planned based on the new connecting point can correspond to the first trajectory.

[0107] In this embodiment, based on the updated surrounding obstacle information, the connection point can be re-determined, the trajectory back to the main road can be planned, and the navigation path to the navigation destination can be further updated. This method avoids frequent activation and deactivation of intelligent driving functions when surrounding obstacles are updated, thus improving the user experience.

[0108] In some embodiments, method 200 can be applied to parking scenarios. The following explanation uses autonomous valet parking as an example.

[0109] For example, when a vehicle first enters a parking lot, it may be parked manually. As the user drives the vehicle within the parking lot, a map of the parking lot (e.g., map #1, which indicates permitted roads along and around path #1) can be automatically constructed based on the vehicle's driving path (e.g., path #1). For instance, when the vehicle parks again in the same parking lot, it can automatically park based on this self-constructed map. Alternatively, different users' vehicles can each construct a local map of the parking lot as they drive within it; the cloud server can then merge these local maps to obtain a global map of the parking lot and distribute it to each user; each user's vehicle can then automatically park based on the map provided by the server.

[0110] In real-world scenarios, vehicles may be in any position. For example, the vehicle's position may be far from the driving path #1. Alternatively, the vehicle may be outside the permitted roads in a parking lot, or even outside the map's coverage area. Taking the implementation of AVP (Autonomous Valet Parking) using a self-constructed map (e.g., map #1) as an example, in traditional solutions, the user needs to drive the vehicle to the road recorded on map #1, or even to the starting point of driving path #1, to activate / enable autonomous valet parking. This severely impacts the user experience. However, in this embodiment, based on surrounding environmental information, a trajectory for the vehicle to return to the main road (e.g., the road where driving path #1 is located, or a road in map #1 that allows return to driving path #1) can be planned. The vehicle is then controlled to return to the main road along this trajectory, enabling automatic parking based on map #1. In other words, the activation / enablement of AVP is no longer limited by the vehicle's position. When the vehicle is not on the main road, it can plan a trajectory to return to the main road based on the surrounding environment information; when the vehicle is on the main road, it can drive towards the navigation destination along the first driving direction of the main road.

[0111] For example, Figure 4 is a flowchart illustrating a control method provided in an embodiment of this application. Method 400 can be understood as an extension or variation of method 200. Method 400 may include the following steps:

[0112] S402 matches the nearest main road to the vehicle and generates a navigation path to the navigation destination.

[0113] In one embodiment, the map used by the vehicle can indicate roads supporting intelligent driving functions, and the main road can be determined based on the map. This map can include at least one of standard definition maps, high-definition maps, and maps autonomously constructed by the vehicle. For example, road information outside underground parking lots can be obtained through a standard definition map. Alternatively, road information inside underground parking lots can be obtained through a map autonomously constructed by the vehicle. Furthermore, the roads surrounding the vehicle's current location can be determined based on the roads indicated by the map. Another example is that a road leading forward to the navigation destination can be determined based on the map and designated as the main road. Yet another example is that the navigation path between the nearest waypoint on the main road and the navigation destination can be determined based on map information.

[0114] S404 determines whether the vehicle's current position meets the conditions for returning to the main road.

[0115] In real-world scenarios, vehicles may be in any position. For example, a vehicle may be in the middle of a main road or outside of a main road.

[0116] In some embodiments, the vehicle's current position is outside the main road. For example, for a certain position, if the vehicle can directly return to the main road from that position using forward / reverse gears, the position can be considered to meet the condition for returning to the main road; if the vehicle cannot directly return to the main road from that position using forward / reverse gears, the position can be considered not to meet the condition for returning to the main road. For another example, the vehicle's ability to directly return to the main road from a certain position using forward / reverse gears can include: using intelligent driving, the vehicle can return to the main road from that position using forward / reverse gears, and after returning to the main road, the angle between the vehicle's longitudinal axis and the first driving direction of the main road is less than or equal to an angle threshold (e.g., 5 degrees, 8 degrees). When the angle between the vehicle's longitudinal axis and the first driving direction of the main road is less than or equal to this angle threshold, the vehicle can drive along the main road towards the navigation destination using intelligent driving. The following examples, in conjunction with the scenarios shown in Figures 6 and 7, illustrate whether the vehicle's position meets the condition for returning to the main road.

[0117] If the conditions for directly returning to the main road are not met, step S406 can be executed; if the conditions for directly returning to the main road are met, step S406 can be skipped, and the vehicle can be controlled to directly return to the main road.

[0118] S406, determine the connection points on the main road.

[0119] The connection points on the main road can be determined based on the vehicle's current pose and the main road information. The distance between the connection point and the nearest road point on the main road to the vehicle's current pose can be less than or equal to a distance threshold, such as 12 meters or 15 meters. The method for determining connection points will be illustrated in Figure 8 below.

[0120] S408, plans the trajectory between the current pose and the connection point.

[0121] For example, the trajectory from the current pose back to the connection point on the main road can be planned based on the surrounding environment information collected by the sensing sensors. For instance, a hybrid A* algorithm, a rapid exploring random trees (RRT) algorithm, or a Dijkstra algorithm can be used to search for the trajectory between the current pose and the connection point.

[0122] In one embodiment, surrounding environmental information can be obtained based on data collected by radar and cameras, and a trajectory to return to the main road can be planned using surrounding empty parking spaces, intersections, and other spatial features. The planning of the trajectory to the return connection point will be illustrated subsequently with reference to the scenario shown in Figure 9.

[0123] S410, European-style space-optimized search trajectory.

[0124] For example, the trajectory obtained in step S408 can be optimized based on Bézier curves and convex optimization algorithms (such as soft constraint methods, hard constraint methods, gradient descent methods, Newton's method, etc.) to obtain an optimized trajectory. Through optimization, a smoother and easier-to-track trajectory can be obtained.

[0125] S412, Update surrounding obstacle information.

[0126] When perception sensors are installed on a vehicle, the information they collect about the surrounding environment is updated as the vehicle moves. Even if obstacles in the environment are stationary, the perceived information about the surrounding environment and obstacles will change at different times as the vehicle moves. Based on the updated obstacle information, it is possible to determine in real time whether there is a risk of collision when the vehicle travels along the originally planned trajectory.

[0127] S414, determine whether to update the connection point and / or replan the trajectory back to the connection point.

[0128] Based on the updated obstacle information, when there is a collision risk while traveling along the originally planned trajectory, the connection points can be updated, and a new trajectory to return to the main road can be planned using the updated connection points. When there is no collision risk while traveling along the originally planned trajectory, there is no need to update the connection points, and the vehicle can be controlled to return to the main road along the originally planned trajectory. The updating of connection points will be illustrated in the following example using the scenario shown in Figure 10.

[0129] S416, track the return to the main road trajectory.

[0130] By tracking the trajectory, the vehicle's motion is controlled, enabling the vehicle to travel precisely along the planned trajectory.

[0131] S418 features real-time lateral motion planning (MOP) based on the main road navigation path.

[0132] A lateral trajectory can be generated based on the navigation path. When cruise conditions are met, the system switches to the lateral trajectory and enters normal cruise mode. The lateral trajectory can be understood as the trajectory along the main road. The starting position of this lateral trajectory may not be on the center line of the main road, but rather there is a certain lateral offset between it and the center line of the main road.

[0133] In this embodiment, by optimizing the MOP line, the transition from the connecting trajectory to the navigation path can be smoothed, improving the user experience.

[0134] S420, longitudinal real-time based on vehicle status, adhesion for speed optimization.

[0135] Adhesion points can be added to the speed optimization model based on the vehicle's real-time speed and the current speed. Adhesion can represent a continuous change in the vehicle's jerk (i.e., the second derivative of speed with respect to time). In other words, the vehicle's jerk changes continuously when switching from a trajectory returning to the main road to driving along the navigation path.

[0136] When acceleration changes continuously, the vehicle's acceleration will not experience sudden changes or abrupt jumps, resulting in a smooth and continuous change in speed. This method avoids the jerky feeling caused by sudden changes or abrupt jumps in vehicle acceleration, achieving a seamless transition and improving the user's riding experience.

[0137] S422, confirm whether you have returned to the main road.

[0138] If the main road has been returned to, step S424 can be executed; if the main road has not been returned to, step S416 can be skipped.

[0139] S424, normal cruise.

[0140] In cruise mode, the vehicle can be controlled to drive intelligently along the navigation route to the destination. After determining the return route to the main road, the vehicle can continue to drive along the navigation route in cruise mode.

[0141] The steps in method 400 are illustrated below with reference to the scenarios shown in Figures 5 to 11.

[0142] Figure 5 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. Figure 5 can serve as an example of a map used by the vehicle. The scenario shown in Figure 5 corresponds to step S402 and can serve as an example of generating a navigation path.

[0143] As shown in Figure 5, the vehicle is outside the road that supports intelligent driving functions. Waypoint 1 can be understood as the waypoint on the main road that is closest to the vehicle's current position, and waypoint 1 can correspond to the second waypoint in method 200. For example, based on map information and the vehicle's current position, the distance between the main road and the vehicle can be determined; waypoint 1 can be determined; and a navigation path can be planned, with the navigation path located on the main road.

[0144] Figure 6 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. In Figure 6, the first direction can represent the direction of the navigation path or the first driving direction of the main road. The scenario shown in Figure 6 can correspond to step S404 and can be used as an example to determine whether the vehicle's pose meets the conditions for returning to the main road.

[0145] In one embodiment, based on whether the lateral offset distance between the vehicle and the main road is less than or equal to a distance threshold (e.g., 5 meters, 6 meters) and whether the angle between the vehicle's longitudinal axis and a first direction is less than or equal to a certain angle threshold (e.g., 50 degrees, 55 degrees), it can be determined whether the vehicle can directly drive to the main road in forward gear. For example, as shown in Figure 6(a), when the lateral offset between vehicle A and the main road is less than or equal to the distance threshold, the vehicle can drive to the main road in forward gear using intelligent driving with a steering angle less than or equal to the maximum steering wheel angle; when the vehicle drives to the main road in forward gear, the vehicle's longitudinal axis can be the first direction (or the angle between it and the first direction is within a small range). As another example, as shown in Figure 6(a), when the angle between the longitudinal axis of vehicle A and the first direction (i.e., angle 1) is less than or equal to the angle threshold, it can be considered that the vehicle can directly return to the main road from this position in forward gear.

[0146] In another embodiment, when a trajectory for returning from the vehicle's current position to the main road can be planned in a preset manner (e.g., using straight lines, arcs, or a combination of arcs and arcs / line segments), it can be assumed that the vehicle can directly return to the main road. For example, as shown in Figures 6(a) and (b), based on the position of vehicle A relative to the main road, trajectories 1 and 2 can be planned in a preset manner. When the vehicle returns to the main road along trajectories 1 and 2, the angle between its longitudinal axis and the first direction is less than a certain angle threshold (e.g., 5 degrees, 6 degrees, 2 degrees), and there is no risk of collision between vehicle A and surrounding obstacles during its journey along trajectories 1 and 2. Trajectory 1 can include a trajectory formed by combining different arcs; trajectory 2 can include a trajectory formed by combining arcs and straight lines. The arc length / radius and the length of the straight lines used in trajectories 1 and 2 can be determined based on the positional relationship between the vehicle and the main road. In the above scenario, it can be assumed that vehicle A can directly return to the main road.

[0147] In some embodiments, given the vehicle's current position, it is impossible to directly return to the main road in forward gear. For example, in the scenario shown in Figure 6(a), when the included angle 1 is greater than the angle threshold, even with the maximum steering wheel angle, the vehicle will not be able to directly return to the main road while in forward gear. In this scenario, it can be considered that the vehicle cannot directly return to the main road from its current position in forward gear. As another example, when there are obstacles in trajectory 1 and trajectory 2, it can be considered that the vehicle cannot directly return to the main road.

[0148] Figure 7 illustrates several scenarios of vehicle poses. The scenario shown in Figure 7 corresponds to step S404 and can be used as an example of a vehicle pose not meeting the conditions for returning to the main road.

[0149] For example, the scenarios shown in Figure 7 can serve as examples of situations where a vehicle cannot directly return to the main road by shifting into forward gear. For instance, when a user wants to activate / enable certain intelligent driving functions in the scenarios shown in Figure 7, in traditional solutions, since vehicle A is not currently on the main road, the user needs to manually drive the vehicle to a position that meets the activation conditions of the intelligent driving function before activating / enabling it. However, in this embodiment, the position of vehicle A in each scenario of Figure 7 can correspond to the first position in method 200. Based on the surrounding environment information, a trajectory (not shown in Figure 7) can be planned to return to the main road from this position, enabling the vehicle to return to the main road along this trajectory in an intelligent driving manner. This allows the user to activate / enable the intelligent driving function at the vehicle's current position, thus expanding the application scenarios of the intelligent driving function.

[0150] In one embodiment, as shown in Figure 7(a), on the one hand, due to the influence of surrounding obstacles, the steering wheel angle that vehicle A can use is smaller than the maximum steering wheel angle; the angle between the longitudinal axis of vehicle A and the first driving direction at waypoint 1 is greater than the available steering wheel angle. On the other hand, when vehicle A travels along a trajectory planned in a preset manner (e.g., by using straight lines, arcs, and arcs / line segments), there is a risk of collision with surrounding obstacles. In this scenario, the vehicle cannot directly return to the main road from its current position. For example, the scenario where vehicle A temporarily stops near an elevator entrance can be illustrated as shown in Figure 7(a).

[0151] In another embodiment, as shown in Figure 7(b), although the distance between vehicle A and waypoint 1 is relatively short, due to the influence of surrounding obstacles (such as other vehicles), the angle between the longitudinal axis of vehicle A and the first driving direction at waypoint 1 is greater than the angle threshold, and vehicle A cannot directly return to the main road from its current pose. For example, the scenario where vehicle A is temporarily parked in a roadside parking space can be illustrated as shown in Figure 7(b).

[0152] In another embodiment, as shown in Figure 7(c), vehicle A deviates from the main road to its current position in order to give way to vehicle B. Even if the angle between the longitudinal axis of vehicle A and the first driving direction at waypoint 1 is less than the angle threshold, and the lateral offset between vehicle A and the main road is less than the distance threshold, due to the influence of surrounding obstacles, it is impossible to plan a trajectory to return to the main road from the previous position by shifting into forward gear based on the method of arc and / or straight line splicing.

[0153] In another embodiment, as shown in Figure 7(d), vehicle A may be in a dead-end road, and the longitudinal axis of vehicle A may be opposite to the first driving direction at waypoint 1. On the one hand, due to the presence of obstacles, vehicle A cannot directly return to the main road from this position in drive. On the other hand, if vehicle A is controlled to return to the main road via waypoint 1 in reverse, after returning to the main road, the angle between its longitudinal axis and the first driving direction of the main road will be large, or even its longitudinal axis may be opposite to the first driving direction of the main road; even after returning to the main road in reverse, the vehicle cannot drive along the main road to the navigation destination in drive.

[0154] In another embodiment, as shown in Figure 7(e), vehicle A may be in a fork in the road. Even if the angle between vehicle A and the first direction of travel at waypoint 1 meets a threshold, and the lateral offset between vehicle A and the main road is less than a distance threshold, it is impossible to directly return to the main road from the current pose by moving forward.

[0155] In another embodiment, as shown in Figure 7(f), waypoint 1 may involve multiple driving directions. For example, due to the impact of obstacles on the detection range of the perception system, or because the intersection is outside the coverage area of ​​the map used by the system, vehicle A may have difficulty knowing the traffic conditions of roads extending in different directions after the intersection in this pose. In traditional solutions, the user may need to manually drive the vehicle through the intersection before activating / enabling certain intelligent driving functions, and it is not possible to directly activate / enable the intelligent driving function in the current pose, nor is it possible to directly pass through the intersection in forward gear using intelligent driving. For another example, in some scenarios where vehicle A deviates from the main road of the area (such as a park, underground parking garage, etc.) after entering the gate, it can be illustrated in Figure 7(f).

[0156] Figure 8 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. The scenario shown in Figure 8 can correspond to step S406 and can be used as an example of determining the connection point on the main road; the scenario shown in Figure 8 can also correspond to step S408 and can be used as an example of trajectory planning.

[0157] As shown in Figure 8, vehicle A is outside the main road. Due to the influence of surrounding obstacles and / or the vehicle's current position, vehicle A cannot directly return to the main road from its current position using forward gear. The connection point can be determined based on the surrounding obstacle information.

[0158] In one embodiment, as shown in Figure 8, by ignoring obstacles around the vehicle, the shortest travel trajectory for the vehicle to travel along the main road from its current pose can be planned, such as trajectory 3. The pose of vehicle A after returning to the main road along trajectory 3 can be denoted as pose B. Based on pose B, reasonable connection points on the main road can be determined. For example, pose B can be translated along the orthogonal direction of the main road to find poses that do not collide with surrounding obstacles, such as pose C. Alternatively, pose B can be moved along the extension direction of the main road to find poses that do not collide with surrounding obstacles, such as poses D and E. In this scenario, the waypoint corresponding to pose B can serve as an example of a first waypoint.

[0159] In another embodiment, in the scenario shown in Figure 8, assuming there are obstacles in the trajectory connecting the current pose and pose C, the connection point on the main road can be determined based on poses D and E. The waypoint where pose C is located can correspond to a candidate waypoint without a reachable trajectory; the waypoints where poses D and E are located can correspond to candidate waypoints with a reachable trajectory. For example, when the connection point on the main road is determined based on pose D, trajectory 4 of the vehicle traveling from the current pose to pose D can be planned; after vehicle A returns to the main road along trajectory 4, it can only travel a short distance along the main road, and then needs to plan the trajectory through the road segments where obstacles 3, 4, and 5 are located again. For another example, to avoid triggering the trajectory planning function multiple times within a short distance, the connection point on the main road can be determined based on pose E; based on the information of surrounding obstacles, trajectory 4 of the vehicle traveling from the current pose to pose D is planned, and trajectory 5 of the vehicle traveling from pose D to pose E is planned. In this scenario, the trajectory formed by trajectory 4 and trajectory 5 can correspond to the first trajectory.

[0160] Figure 9 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. The scenario shown in Figure 9 can correspond to step S408 and can serve as an example of trajectory planning.

[0161] In one embodiment, as shown in Figure 9(a), when there is an empty parking space on one side of the vehicle, the empty parking space in the environment can be used to plan the trajectory for the vehicle to return from its current position to the connection point. After traveling along this trajectory to the connection point, the vehicle can travel along the navigation path.

[0162] In another embodiment, as shown in Figure 9(b), when there are empty parking spaces on both sides of the vehicle, the empty parking spaces in the environment can be fully utilized to plan the trajectory of the vehicle returning from its current position to the connection point.

[0163] In another embodiment, as shown in Figure 9(c), when there are no empty parking spaces around the vehicle or the empty parking spaces are far away, the trajectory from the current position back to the connection point can be planned by multiple rubbing of the parking space.

[0164] In this embodiment of the application, by utilizing empty parking spaces in the surrounding environment to plan the trajectory of the vehicle returning from its current position to the connection point, the number of gear shifts when the vehicle travels along the trajectory can be reduced, and the length of the trajectory from the vehicle's current position to the connection point can be shortened as much as possible.

[0165] Figure 10 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. The scenario shown in Figure 10 can correspond to step S412, and can be used as an example of updating obstacle information as the vehicle moves; it can also correspond to step S414, and can be used as an example of updating connection points.

[0166] As shown in Figure 10(a), multiple vehicles located in the parking area are outside the perception range of vehicle A. Connection point 1 can serve as the connection point for vehicles to return to the main road, and correspondingly, connection point 1 can serve as the starting point of navigation path #1.

[0167] As shown in Figure 10(b), the planned trajectory from position #1 back to connection point 1 can include trajectory 6 and trajectory 7. Vehicle A can drive in reverse along trajectory 6 from position #1 to position #2, and in drive along trajectory 7 from position #2 to connection point 1. In this scenario, the trajectory formed by trajectory 6 and trajectory 7 can correspond to the first trajectory in method 200; trajectory 7 can correspond to the connection trajectory.

[0168] As shown in Figure 10(c), during the travel of vehicle A along trajectory 6, the vehicle's perception range changes with its movement. Vehicles stationary in the parking area will enter the perception range of vehicle A. For example, during the travel of vehicle A along trajectory 6, as the perceived surrounding obstacles are updated, if it is determined that continuing along trajectory 6 will pose a risk of collision with surrounding obstacles, the connection point on the main road can be updated from connection point 1 to connection point 2; correspondingly, the navigation path on the main road to the navigation endpoint can be updated from navigation path #1 to navigation path #2.

[0169] As shown in Figure 10(d), based on the updated obstacle information, the connection point on the main road can be updated from connection point 1 to connection point 2, and trajectory 8, which takes place from position #3 back to the main road via connection point 2, can be planned. For the intelligent driving process from position #1 to the navigation endpoint, vehicle A can drive in reverse along trajectory 6 from position #1 to position #3, and can then drive from position #3 along trajectory 8 with multiple gear changes to reach connection point 2; furthermore, it can drive along navigation path #2 to the navigation endpoint. In this scenario, the actual trajectory of vehicle A returning from position #1 to the main road can include the portion of trajectory 6 between position #1 and position #3, as well as trajectory 8. In this scenario, the portion of trajectory 6 between position #1 and position #3 can correspond to the third trajectory in method 200; for the updated surrounding environment, the vehicle's pose at position #3 can correspond to the first pose, and correspondingly, trajectory 8 can correspond to the first trajectory, with the last segment of trajectory 8 corresponding to the connecting trajectory.

[0170] Figure 11 is a schematic diagram of another intelligent driving scenario provided by an embodiment of this application. The scenario shown in Figure 11 can correspond to steps S418 and S420.

[0171] As shown in Figure 11, the connecting trajectory can be understood as the last segment of the trajectory before the connecting point in the trajectory from the vehicle's current position back to the connecting point.

[0172] When vehicle A is moving along the connecting trajectory, the cruise conditions can be determined based on the vehicle's position and posture.

[0173] In one embodiment, as shown in Figure 11(b), during the process of returning to the connection point along the connection trajectory, when the lateral offset between the vehicle and the centerline of the main road is less than or equal to a first threshold, and / or the angle between the longitudinal axis direction of the vehicle and the navigation direction of the main road at the connection point is less than a second threshold, the cruise conditions can be considered met. At this time, the vehicle can be controlled to switch to cruise along the navigation path.

[0174] In another embodiment, as shown in Figure 11(c), during the process of the vehicle returning from the connecting trajectory to the main road and cruising on the main road, the vehicle speed at the adhesion point can be planned so that the vehicle speed changes continuously before and after the trajectory switch. The vehicle's acceleration and jerk at the adhesion point can also be planned so that the vehicle's acceleration and jerk change continuously before and after the adhesion point, resulting in a smooth and continuous change in vehicle speed (as shown in Figure 11(c)). The time of the adhesion point can correspond to the first time point in method 200.

[0175] The methods provided by the embodiments of this application have been described in detail above with reference to Figures 2 to 11. The apparatus provided by the embodiments of this application will now be described in detail below with reference to Figures 12 and 13. The descriptions of the apparatus embodiments correspond to the descriptions of the method embodiments; therefore, any content not described in detail can be found in the method embodiments above.

[0176] For example, FIG12 shows a schematic block diagram of an apparatus 2000 provided in an embodiment of the present application, which may include an acquisition unit 2010 and a processing unit 2020.

[0177] When the device 2000 is used to execute the method 200 in FIG2, the acquisition unit 2010 can be used to execute step S210 in the method 200; the processing unit 2020 can be used to execute steps S220 and S230 in the method 200.

[0178] Specifically, the acquisition unit 2010 can be used to: acquire information about the vehicle's surrounding environment. The processing unit 2010 can be used to: plan a first trajectory for the vehicle to return from its first pose to the main road based on the information about the vehicle's surrounding environment; and control the vehicle to return to the main road along the first trajectory.

[0179] For example, the device 2000 may be a computing platform 150 (such as an autonomous driving domain controller), or it may be a processor or chip in an autonomous driving domain controller, or it may be a vehicle or intelligent driving system equipped with an autonomous driving domain controller.

[0180] In some embodiments, the processing unit 2020 may be used to: determine the connection point on the main road based on the surrounding environment information; and plan the first trajectory from the first pose back to the main road based on the connection point.

[0181] In other embodiments, the processing unit 2020 can be used to: control the vehicle to switch from the connecting trajectory to travel along the main road when the switching conditions are met during the vehicle's travel along the connecting trajectory.

[0182] In some other embodiments, the switching condition is met at a first moment within a first time period. The processing unit 2020 can be used to control the vehicle speed to change continuously within the first time period.

[0183] In some other embodiments, the vehicle's acceleration changes continuously at the first moment.

[0184] In other embodiments, the processing unit 2020 may be used to: determine connection points based on candidate waypoints with reachable trajectories.

[0185] In some embodiments, candidate waypoints with reachable trajectories include a first candidate waypoint. The processing unit 2020 can be used to: plan a first candidate trajectory from the first pose back to the main road based on the first candidate waypoint; and determine whether to designate the first candidate waypoint as a connection point based on the cost of the first candidate trajectory.

[0186] It should be understood that the division of units in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. All units of the above device can be implemented entirely through processor-invoked software, entirely through hardware circuits, or partially through processor-invoked software with the remaining parts implemented through hardware circuits.

[0187] In a specific implementation, the acquisition unit 2010 can be implemented by at least one processor or processor-related circuitry, and the processing unit 2020 can be implemented by at least one transceiver or transceiver-related circuitry. In one example, one or more processors can acquire information about the vehicle's surrounding environment. In another example, one or more processors can plan a first trajectory from the first pose back to the main road. Exemplarily, in a specific implementation, the device 2000 can be an intelligent driving device equipped with an autonomous driving domain controller, or a chip or processor located within the autonomous driving domain controller.

[0188] For example, FIG13 is a schematic block diagram of another device 3000 provided in an embodiment of this application. The device 3000 may include a processor 3010, an interface circuit 3020, and a memory 3030. The processor 3010, the interface circuit 3020, and the memory 3030 are connected via internal connection paths. The memory 3030 is used to store instructions, and the processor 3010 is used to execute the instructions stored in the memory 3030, so that the interface circuit 3020 can receive / send some parameters. Optionally, the memory 3030 may be coupled to the processor 3010 via an interface, or it may be integrated with the processor 3010.

[0189] It should be noted that the aforementioned interface circuit 3020 may include, but is not limited to, transceiver devices such as input / output interfaces, to enable communication between device 3000 and other devices or communication networks. For example, the interface circuit 3020 can be used to acquire surrounding environmental information and / or transmit control commands to cause the vehicle to travel along a planned trajectory / path.

[0190] This application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to execute any of the method embodiments in Figures 2 to 11 above, and any possible implementation thereof.

[0191] This application also provides a computer-readable storage medium storing program code or instructions that, when executed by a computer's processor, cause the processor to implement any of the method embodiments in Figures 2 to 11 above, and any possible implementation thereof.

[0192] This application also provides a chip, including a circuit, for executing any of the method embodiments in Figures 2 to 11 above, and any possible implementation thereof.

[0193] This application also provides an intelligent driving system, which may include the above-described device 2000 or 3000.

[0194] This application also provides an intelligent driving device, which may include the above-described device 2000 or 3000, or the above-described intelligent driving system.

[0195] For example, the intelligent driving device can be a vehicle. The vehicle involved in this application embodiment is a vehicle in a broad sense, which can be a means of transportation (such as commercial vehicles, passenger cars, motorcycles, flying cars, trains, etc.), industrial vehicles (such as forklifts, trailers, tractors, etc.), engineering vehicles (such as excavators, bulldozers, cranes, etc.), agricultural equipment (such as lawnmowers, harvesters, etc.), amusement equipment, toy vehicles, etc. For example, the vehicle in this application can include pure electric vehicles (pure electric vehicle / battery electric vehicle, pure EV / battery EV), hybrid electric vehicles (HEV), range-extended electric vehicles (REEV), plug-in hybrid electric vehicles (PHEV), or new energy vehicles (NEV), etc.

[0196] Those skilled in the art will 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, or a combination of computer software and electronic hardware. 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 implementation should not be considered beyond the scope of this application.

[0197] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0198] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0199] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0200] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0201] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0202] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent driving, characterized in that, include: Obtain information about the vehicle's surrounding environment; Based on the surrounding environment information, a first trajectory is planned for the vehicle to return to the main road from the first position, where the first position is outside the main road, and the main road includes roads that support intelligent driving functions and can be driven to the navigation destination. Control the vehicle to return to the main road along the first trajectory.

2. The method according to claim 1, characterized in that, The step of planning the first trajectory for the vehicle to return to the main road from the first position based on the surrounding environment information includes: Based on the surrounding environment information, determine the connection points on the main road; Based on the connection point, plan the first trajectory.

3. The method according to claim 1 or 2, characterized in that, The first trajectory includes a connecting trajectory ending at the connecting point on the main road, in which the vehicle uses either forward or reverse gear; The method further includes: During the vehicle's travel along the connecting trajectory, when the switching conditions are met, the vehicle is controlled to switch to travel along the main road. The switching conditions include at least one of the following: the second pose of the vehicle meets a preset condition; the lateral distance between the second pose of the vehicle and the centerline of the main road is less than or equal to a first threshold; and the angle between the longitudinal axis of the vehicle and the first driving direction of the main road is less than or equal to a second threshold.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes satisfying the switching condition at the first moment within the first time period: During the first time period, the vehicle speed is controlled to vary continuously.

5. The method according to claim 4, characterized in that, At the first moment, the vehicle's acceleration changes continuously.

6. The method according to any one of claims 2 to 5, characterized in that, The first waypoint is the endpoint of the shortest travel trajectory for the vehicle to return to the main road from the first position when there are no surrounding obstacles. Determining the connection points on the main road includes: The first waypoint is translated along the extension direction of the main road and / or the orthogonal direction of the extension direction to determine candidate waypoints with reachable trajectories. When the vehicle returns to the main road from the first pose along the reachable trajectory, there is no risk of collision with surrounding obstacles. The connection point is determined based on the candidate waypoints with reachable trajectories.

7. The method according to claim 6, characterized in that, The candidate waypoints include a first candidate waypoint, and determining the connection point based on the candidate waypoints with reachable trajectories includes: Plan the first candidate trajectory from the first pose to the first candidate waypoint; Based on the cost of the first candidate trajectory, it is determined whether to use the first candidate waypoint as the connection point. The cost of the first candidate trajectory is determined based on at least one of the length of the first candidate trajectory, the number of gear shifts during the vehicle's journey along the first candidate trajectory, and the number of steering direction changes during the vehicle's journey along the first candidate trajectory.

8. The method according to any one of claims 2 to 7, characterized in that, The second waypoint is the waypoint on the main road that is closest to the first pose, and the connection point is located within the first range corresponding to the second pose.

9. The method according to claim 7 or 8, characterized in that, The method further includes: Based on the surrounding environment information, a first vacant area is determined. The first vacant area is used to plan the first candidate trajectory. The first vacant area includes at least one of the following: vacant parking spaces, vacant intersections, and wide lanes.

10. The method according to any one of claims 2 to 9, characterized in that, The method further includes: It is determined that the first pose does not meet the preset conditions; The preset conditions include: when the vehicle returns to the main road from the first position along a trajectory planned in a preset manner, there is no risk of collision with surrounding obstacles.

11. The method according to any one of claims 1 to 10, characterized in that, The first driving direction of the main road is from the first waypoint on the main road closest to the vehicle to the navigation destination.

12. A device for intelligent driving, characterized in that, include: The acquisition unit is used to: acquire information about the vehicle's surrounding environment; The processing unit is configured to: plan a first trajectory for the vehicle to return from a first pose to the main road based on the surrounding environment information, wherein the first pose is outside the main road and the main road includes roads that support intelligent driving functions and can be driven to the navigation destination; Control the vehicle to return to the main road along the first trajectory.

13. The apparatus according to claim 12, characterized in that, The processing unit is used for: Based on the surrounding environment information, determine the connection points on the main road; Based on the connection point, plan the first trajectory.

14. An apparatus, characterized in that, The device includes at least one processor coupled to at least one memory, the at least one processor being configured to execute a computer program or instructions stored in the at least one memory to cause the device to perform the method as described in any one of claims 1 to 11.

15. An intelligent driving system, characterized in that, Includes the apparatus as described in any one of claims 12 to 14.

16. A computer-readable storage medium, characterized in that, It stores instructions or program code thereon, which, when executed by a processor, cause the processor to implement the method as described in any one of claims 1 to 11.

17. A computer program product, characterized in that, The computer program product includes: computer program code, which, when executed, implements the method as described in any one of claims 1 to 11.

18. A chip, characterized in that, The circuit includes a communication interface for receiving information from other devices and inputting it into the circuit, and / or the communication interface for sending information in the circuit to other devices, the circuit being used to perform the method as described in any one of claims 1 to 11.

19. A vehicle, characterized in that, Includes the device as described in any one of claims 12 to 14, or the intelligent driving system as described in claim 15.

Images