CN115892075B - Trajectory planning method, automatic driving device, and computer storage medium - Google Patents
Trajectory planning method, automatic driving device, and computer storage medium Download PDFInfo
- Publication number
- CN115892075B CN115892075B CN202310018461.0A CN202310018461A CN115892075B CN 115892075 B CN115892075 B CN 115892075B CN 202310018461 A CN202310018461 A CN 202310018461A CN 115892075 B CN115892075 B CN 115892075B
- Authority
- CN
- China
- Prior art keywords
- automatic driving
- space
- driving equipment
- traffic
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 230000008569 process Effects 0.000 claims description 23
- 238000006073 displacement reaction Methods 0.000 claims description 21
- 230000008447 perception Effects 0.000 claims description 18
- 238000009499 grossing Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000004888 barrier function Effects 0.000 claims description 5
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 2
- 230000008030 elimination Effects 0.000 claims description 2
- 238000003379 elimination reaction Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 10
- 238000004891 communication Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 13
- 230000002829 reductive effect Effects 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000003068 static effect Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 238000000605 extraction Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 4
- 238000002955 isolation Methods 0.000 description 4
- 239000004973 liquid crystal related substance Substances 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Abstract
The embodiment of the application provides a track planning method, automatic driving equipment and a computer storage medium, wherein the track planning method comprises the following steps: determining a first running space which can be driven for the automatic driving equipment according to map information corresponding to the environment where the automatic driving equipment is located; determining a second passing space for the automatic driving equipment from the first passing space according to the sensing information acquired by the automatic driving equipment; determining a target passing space from the second passing space; and planning the track of the automatic driving equipment according to the target passing space. According to the embodiment of the application, the safe and efficient running track can be planned on the vehicle-mounted computing platform with limited computing power while the algorithm is ensured to be light.
Description
Technical Field
The embodiment of the application relates to the technical field of automatic driving, in particular to a track planning method, automatic driving equipment and a computer storage medium.
Background
In the field of automatic driving, a driving track is a curve which consists of a series of coordinate position points expected to be reached by a vehicle and information such as time, speed, acceleration and the like when the vehicle is expected to reach the coordinate position, and the curve comprises two dimensions of time and space.
In order to ensure the safety and high efficiency of the planned driving track, the automatic driving equipment is generally required to have stronger hardware performance so as to support an algorithm frame for carrying out the planning of the driving track. However, for those small and medium-sized autopilot devices, which do not have very high hardware performance, it is difficult to support a similar algorithmic framework due to the limited computing power of their onboard computing platforms.
Therefore, on the vehicle-mounted computing platform with limited computing power, the problem of planning a safe and efficient driving track while ensuring the light weight of an algorithm is urgent to be solved.
Disclosure of Invention
In view of the foregoing, embodiments of the present application provide a trajectory planning scheme to at least partially solve the above-mentioned problems.
According to a first aspect of an embodiment of the present application, there is provided a trajectory planning method, including: determining a first running space which can be driven for the automatic driving equipment according to map information corresponding to the environment where the automatic driving equipment is located; determining a second passing space for the automatic driving equipment from the first passing space according to the sensing information acquired by the automatic driving equipment; determining a target passing space from the second passing space; and planning the track of the automatic driving equipment according to the target passing space.
According to a second aspect of embodiments of the present application, there is provided an automatic driving apparatus including: a sensor and a processor; wherein: the sensor is used for collecting environmental data of the environment where the automatic driving equipment is located and generating perception information; the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method according to the first aspect based on the sensed information.
According to a third aspect of embodiments of the present application, there is provided a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the method according to the first aspect.
According to a fourth aspect of embodiments of the present application, there is provided a computer program product comprising computer instructions for instructing a computing device to perform the operations corresponding to the method according to the first aspect.
According to the track planning scheme provided by the embodiment of the application, when the traffic space is determined for the automatic driving equipment, the first traffic space which can be driven is determined by a static obstacle removing mode based on map information; then, based on the perception information acquired by the automatic driving equipment, excluding the non-traffic spaces indicated by the perception information from the first traffic space, so as to obtain a second traffic space; and then determining the target traffic space from the second traffic space. On one hand, the process does not need to rely on a complex machine learning model to carry out space decision processing, so that the algorithm complexity is greatly reduced; on the other hand, the trajectory planning of the automatic driving device is performed in the determined target traffic space, namely, the unreasonable traffic space is effectively eliminated, and the data calculation amount of the space decision is further reduced. Therefore, the track planning scheme of the embodiment of the application can be suitable for a vehicle-mounted computing platform with limited calculation force, and a safe and efficient running track can be planned while the light weight of an algorithm is ensured.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present application, and other drawings may also be obtained according to these drawings for a person having ordinary skill in the art.
FIG. 1 is a schematic diagram of an exemplary system to which the trajectory planning scheme of embodiments of the present application may be applied;
FIG. 2 is a flow chart of steps of a trajectory planning method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a Cartesian coordinate system and a Frenet coordinate system according to the embodiment shown in FIG. 2;
FIG. 4 is a schematic view of a first traffic space extraction in the embodiment of FIG. 2;
FIG. 5 is a schematic diagram of a process for constructing a kinematic estimation model in a second through-space extraction process in the embodiment shown in FIG. 2;
FIG. 6 is a schematic diagram of a process for constructing a kinematic estimation model in another second through-space extraction process in the embodiment shown in FIG. 2;
FIG. 7 is a schematic diagram of a forward search during a second through-space extraction process in the embodiment of FIG. 2;
FIG. 8 is a schematic diagram of end position clipping in a second through-space extraction process in the embodiment of FIG. 2;
FIG. 9 is a schematic diagram of a backward search in a second through-space extraction process in the embodiment of FIG. 2;
FIG. 10 is a schematic view of a traffic space smoothing process in the embodiment of FIG. 2;
FIG. 11 is a schematic diagram of an overall trajectory planning process in the embodiment of FIG. 2;
fig. 12 is a schematic structural view of an automatic driving apparatus according to an embodiment of the present application.
Detailed Description
In order to better understand the technical solutions in the embodiments of the present application, the following descriptions will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the embodiments of the present application shall fall within the scope of protection of the embodiments of the present application.
Embodiments of the present application are further described below with reference to the accompanying drawings of embodiments of the present application.
Fig. 1 shows an exemplary system to which the trajectory planning method of the embodiments of the present application is applied. As shown in fig. 1, the system 100 may include a cloud service 102, a communication network 104, and/or one or more autopilot devices 106, which are illustrated in fig. 1 as a plurality of autopilot devices.
Cloud server 102 may be any suitable device for storing information, data, programs, and/or any other suitable type of content, including, but not limited to, distributed storage system devices, server clusters, computing cloud server clusters, and the like. In some embodiments, cloud server 102 may perform any suitable functions. For example, in some embodiments, cloud service 102 may be used to store various data sent by autopilot device 106. As an alternative example, in some embodiments, cloud service 102 may also send data requested by the request to autopilot device 106 upon request by autopilot device 106.
In some embodiments, the communication network 104 may be any suitable combination of one or more wired and/or wireless networks. For example, the communication network 104 can include any one or more of the following: the internet, an intranet, a Wide Area Network (WAN), a Local Area Network (LAN), a wireless network, a Digital Subscriber Line (DSL) network, a frame relay network, an Asynchronous Transfer Mode (ATM) network, a Virtual Private Network (VPN), and/or any other suitable communication network. The autopilot device 106 can be coupled to the communication network 104 via one or more communication links (e.g., communication link 112), which communication network 104 can be linked to the cloud service 102 via one or more communication links (e.g., communication link 114). The communication link may be any communication link suitable for transferring data between the autopilot device 106 and the cloud service 102, such as a network link, dial-up link, wireless link, hardwired link, any other suitable communication link, or any suitable combination of such links.
Based on the above system, the embodiment of the application provides a track planning scheme, so as to be mainly applied to an automatic driving device of a vehicle-mounted computing platform with limited bearing capacity. It will be apparent to those skilled in the art that in practical applications, the solution of the embodiments of the present application is equally applicable to autopilot devices with higher computing power. The trajectory planning scheme of the present application is described below by way of examples.
Referring to fig. 2, a flowchart of steps of a trajectory planning method according to an embodiment of the present application is shown.
The track planning method of the present embodiment includes the following steps:
step S202: and determining a first running space which can be driven for the automatic driving equipment according to the map information corresponding to the environment where the automatic driving equipment is located.
In the embodiment of the application, the automatic driving device is a device with an automatic driving function, including but not limited to a full-automatic driving device and an automatic driving device with manual automatic driving and automatic driving functions, such as a logistics distribution vehicle, a service robot and the like.
When track planning is performed for the automatic driving device, map information and sensing information acquired by the automatic driving device are generally needed, but in a traditional mode, track planning is performed based on the sensing information, the map information is only used for providing road information, and the mode leads to complex algorithm framework and higher calculation force requirement. In the embodiment of the application, the map information is first used to determine the first available traffic space for the autopilot device.
The map information not only comprises information such as roads, lanes, lane lines and the like, but also comprises information such as road foundation elements such as road edges, railings, isolation belts, flower beds, road piles and the like, such as position coordinates, topological relations, altitude information and the like. These curbs, railings, isolation belts, flower beds, road piles, etc. are typically static, fixed, but are obstacles that prevent passage. Therefore, the vehicle can be regarded as a fixed passage obstacle, and the space formed by the elements can be firstly eliminated, so that a first movable and static passage space is determined.
That is, a fixed passing obstacle in an environment can be determined according to map information corresponding to the environment in which the automatic driving device is located; and removing the fixed traffic barrier, and determining a first traffic space which can be driven for the automatic driving equipment according to the removed map information. The first traffic space can be generated offline in advance to further save the calculation force of the automatic driving equipment for real-time track planning. However, the method is not limited thereto, and the determination of the first traffic space may be performed when the automatic driving apparatus is needed, which may also be applied to the solution of the embodiment of the present application, and may still save a certain amount of effort compared to the conventional method.
For example, in one possible manner, map information of a certain area in front of the driving direction of the automatic driving device may be obtained according to the current position of the automatic driving device, and then, a fixed passing obstacle therein may be excluded according to the map information of the partial area, so as to obtain the first passing space. For another example, in another possible manner, the area information in front of the driving direction of the automatic driving apparatus may be acquired according to the current position of the automatic driving apparatus, and the first traffic space generated in advance corresponding to the area information may be determined.
Illustratively, taking the example of the autopilot device determining its first traffic space under Frenet (Frenet in Freler) coordinate system, a schematic representation of a first traffic space extraction is shown in FIG. 4.
For the convenience of explanation of this example, a cartesian coordinate system and a Frenet coordinate system that are often used for trajectory planning will be explained below with reference first to fig. 3. First, as shown on the left side of fig. 3, a conventionally used cartesian coordinate system, also called world coordinate system, is generally represented by a horizontal axis X and a vertical axis Y, and the position of an object in the cartesian coordinate system is described using (X, Y) coordinates. However, in the trajectory planning of the autopilot, even if the position (x, y) of the autopilot is known, its relationship to the road cannot be known. The Frenet coordinate system describes the position of the autopilot relative to the road, as shown on the right side of FIG. 3, with the distance of the autopilot along the road as the vertical axis S and the displacement of the autopilot from the longitudinal line as the horizontal axis D. In this way, it is ensured that at each point of the road, both the horizontal and vertical axes are vertical, the ordinate may represent the distance travelled by the autopilot in the road, and the abscissa may represent the distance that the autopilot deviates from the centre line. Based on these two coordinate systems, the person skilled in the art can, if necessary, use conventional conversion means to convert corresponding data between these two coordinate systems, the specific conversion algorithm not being described in detail here.
Based on this, as shown in fig. 4, according to map information (such as high-precision map information) and geometric parameters (such as vehicle width, etc.) of the own vehicle, the space where the own vehicle cannot pass, such as isolation zones, flower beds, too narrow road piles, etc., is excluded, and the space geometric information in the map where the own vehicle can normally travel without considering the obstacle determined by perception is extracted. If the space geometry information is information in a Cartesian coordinate system, the space geometry information can be converted into space geometry information in a Frenet coordinate system. In Frenet coordinate system, the space geometric information can be expressed as the transverse coordinate range of the space which can be driven on each ordinate s
, 
) As indicated by the solid bi-directional lines with arrows in fig. 4.
In this way, a first, stationary passage space that can be traveled can be determined for the autopilot system.
Step S204: and determining a second passing space for the automatic driving equipment from the first passing space according to the sensing information acquired by the automatic driving equipment.
In actual road travel, there are not only fixed traffic obstacles shown in the map but also some temporary obstacles such as other traffic devices (other motor vehicles, non-motor vehicles, pedestrians, temporarily placed obstacles, etc.), or some non-temporary non-passable spaces such as too narrow roads, etc. Such information needs to be further perceived by the perception information collected by the autopilot device (including but not limited to camera-collected images, radar-collected point cloud data, etc.), and based thereon, further spatial range exclusion.
That is, it is necessary to determine temporary traffic obstacles or non-drivable traffic spaces with widths smaller than a preset width threshold in an environment where the automatic driving device is located according to the sensing information collected by the automatic driving device; removing temporary traffic barriers or non-drivable traffic spaces from the first traffic space; and determining a second passing space for the automatic driving equipment according to the elimination result. The preset width threshold may be flexibly set by a person skilled in the art according to actual situations, and is generally set to a width that most of automatic driving devices cannot pass through, or is set according to the width of the vehicle. By the method, the space which cannot be passed in the actual driving process is further eliminated, and the accuracy of the determined passing space which can be driven is ensured.
When the second through space determination is specifically performed, in a feasible manner, a current reference line corresponding to the automatic driving device can be determined; determining a matched kinematic estimation model according to the curvature of the current reference line; and determining a second passing space for the automatic driving equipment from the first passing space according to the determined kinematic estimation model and the perception information acquired by the automatic driving equipment. By means of a kinematic estimation model matching the curvature of the current reference line, a possible passage space can be determined more accurately for the autopilot device.
The reference line is a smooth line which can be dynamically generated to guide the automatic driving equipment to run according to the high-precision map data, the self-vehicle positioning information and the starting point and end point coordinate information of the track planning task. Although the line is mostly curved, there are also cases where it is partly straight, i.e. the curvature is 0. Different kinematic estimation models may be employed to determine the likely traffic space of the autopilot.
To this end, in one possible way, determining a matching kinematic estimation model according to the curvature of the current reference line may be implemented as: determining the relation between the transverse displacement and the longitudinal displacement of the automatic driving equipment under the Frenet coordinate system according to the current position of the automatic driving equipment and the curvature of the current reference line; a kinematic estimation model is determined that matches the curvature of the current reference line based on the relationship.
For example, a current reference line of the autopilot device may be determined from a current location of the autopilot device; judging whether the curvature of the current reference line is 0; if the curvature of the current reference line is 0, determining that the relation between the lateral displacement and the longitudinal displacement of the automatic driving device under the Frenet coordinate system is as follows: the lateral displacement is less than or equal to the product of the longitudinal displacement and the spatial expansion coefficient of the autopilot device in a cartesian coordinate system. If the curvature of the current reference line is not 0, the relation between the transverse displacement and the longitudinal displacement of the automatic driving device under the Frenet coordinate system can be determined according to the sine value and the cosine value of the product of the longitudinal displacement and the curvature.
For example, as shown in fig. 5, when the curvature of the reference line is 0, the Frenet coordinate system has a linear correspondence with the cartesian coordinate system, and the kinematic estimation model corresponding to the curvature of 0 may be expressed as:
. Wherein, the liquid crystal display device comprises a liquid crystal display device,
the distance of each change of the ordinate under the Frenet coordinate system is represented, namely the distance of the automatic driving equipment moving along the longitudinal direction; k represents the spatial expansion coefficient of the automatic driving equipment under the Cartesian coordinate system, and can be adjusted according to specific parameters (such as turning radius) of the automatic driving equipment;
representing the distance each time the abscissa changes, i.e. the lateral distance that the autopilot device may reach while moving in the longitudinal direction, under the Frenet coordinate system. Since in the case of curvature 0, the directions along the abscissa are different
Should be equal in theory, therefore, in this way, unify
Representing the lateral movement distances in different directions along the abscissa.
However, when the curvature of the reference line is not 0, as shown in fig. 6, the Frenet coordinate system does not have a linear correspondence with the cartesian coordinate system. In this case, the kinematic estimation model corresponding to curvature other than 0 can be expressed as:
where kappa denotes the curvature of the reference line, when the curvature of the reference line is not constant, an approximate estimate can be made using the curvature of the reference line at the middle position. As shown in the figure 6 of the drawings, 
A left lateral offset position representing the current position of the autopilot,
a right lateral offset position representing a current position of the autopilot device;
the distance of each change of the ordinate under the Frenet coordinate system is represented, namely the distance of the automatic driving equipment moving along the longitudinal direction;
the distance of each change of the left-hand abscissa under the Frenet coordinate system, namely the distance which can be reached by the left-hand driving device when the automatic driving device moves along the longitudinal direction;
the distance of each change of the right-direction abscissa under the Frenet coordinate system, namely the distance possibly reached by the right-direction driving device when the automatic driving device moves along the longitudinal direction; k represents the spatial expansion coefficient of the autopilot in a cartesian coordinate system, which can be adjusted according to specific parameters of the autopilot (e.g. turning radius).
From the above, the above formula is also applicable when the curvature of the reference line is 0, i.e
The corresponding model may be equally applicable to the case where the curvature is 0. Nevertheless, when the curvature is 0, then
In the way of (a), the amount of computation can be greatly reduced.
On the basis of determining the kinematic estimation model, further, a second traffic space can be determined for the automatic driving device from the first traffic space according to the determined kinematic estimation model and perception information acquired by the automatic driving device. For example, a forward search may be performed in the first traffic space along the forward direction of the autopilot device based on the determined kinematic estimation model and the perceived information collected by the autopilot device to obtain at least one second traffic space.
The forward search is started from the vehicle position, and according to the curvature of the reference line and the kinematic estimation model matched with the curvature, the transverse coordinate range which can be reached by the vehicle at the next longitudinal coordinate position is gradually searched along the increasing direction of the longitudinal coordinate s. An exemplary forward search is schematically shown in fig. 7, and as can be seen from fig. 7, 3 driving channels are finally searched out through forward search, wherein the channel 1 is a dead-end scene because the forward obstacle blocks the search from being terminated in advance; the channel 2 and the channel 3 can enable the self-vehicle to normally run, and the difference is that the rear section of the channel 2 runs on a motor vehicle lane on the left side of the flower bed, and the rear section of the channel 3 runs on a non-motor vehicle lane on the right side of the flower bed. In this case, the travel spaces corresponding to the lane 2 and the lane 3 can be regarded as the second travel space.
It can be seen that the second through space is composed of a series of consecutive lateral coordinate ranges of the longitudinal coordinates s, and is characterized in that the vehicle can plan one or more (two or more) paths from the longitudinal coordinate positions to the longitudinal coordinate positions in the space. Therefore, the method further eliminates the space which cannot pass such as road blockage, over-narrow and the like in the acquired static first passing space under the Frenet coordinate system by combining the positioning information of the own vehicle under the condition of considering the perceived and detected obstacle, and searches a running channel (running space) which can normally pass by the own vehicle, namely a second passing space by a kinematic estimation model under the condition of considering the constraint of the own vehicle kinematics. Because the traditional kinematic models aiming at the automatic driving equipment are mostly the acarman models, the model is inconvenient to directly use for estimating the traffic space, the novel kinematic estimation model is provided in the scheme of the embodiment of the application, and the data calculation amount and the calculation power consumption are greatly saved on the basis of ensuring that the traffic space estimation can be effectively performed.
Step S206: and determining the target traffic space from the second traffic space.
While the second traffic space has been given one or more traffic spaces that can be traveled, different driving devices may have different driving scenarios for the autopilot device, and forward search is a process of gradually expanding the lateral range, often to an end point where the lateral range is too large. Therefore, it is also necessary to determine an appropriate target traffic space from the second traffic space according to the specific driving scenario of the automatic driving apparatus.
Based on this, in one possible way, this step can be implemented as: performing space clipping on at least one second through space according to the expected end position of the automatic driving equipment; and obtaining a target passing space according to the cutting result.
For example, the lateral range of the second passing space may be cut at the end position as required, taking the automatic driving device as a logistics vehicle for example, and being a low-speed vehicle, the automatic driving device is more suitable for walking on a non-motor vehicle lane or a rightmost motor vehicle lane, as shown in fig. 8, the lateral range of the end position may be cut according to the principle, so as to obtain two target passing spaces, and two passing spaces corresponding to the range shown by the thick solid line with the end range being a double-headed arrow in fig. 8 respectively.
However, in order to ensure the accuracy of determining the target traffic space, in one possible manner, when the target traffic space is obtained according to the clipping result, at least one candidate traffic space (for example, two traffic spaces corresponding to a range shown by a thick solid line with an end point range being a double-headed arrow in fig. 8) may be obtained according to the clipping result; then, according to the determined kinematic estimation model and the perception information acquired by the automatic driving equipment, in at least one candidate passing space, backward searching is carried out along the backward direction of the automatic driving equipment; and obtaining a final target passing space according to the backward search result.
In order to ensure that the front and rear coordinate positions can be reached by traveling in the same traffic space, a backward search in the opposite direction to the forward search is required. For example, as shown in fig. 9, for the traffic space corresponding to the lane 2 and the traffic space corresponding to the lane 3, from the end position, according to the determined kinematic estimation model, the next longitudinal coordinate position and the lateral coordinate range that the own vehicle can reach are searched gradually along the direction in which the longitudinal coordinate s decreases. It should be noted that the lateral coordinate range cannot exceed the lateral coordinate range obtained by the forward search. In this example, through backward search, it may be determined that the traffic space corresponding to the channel 2 and the traffic space corresponding to the channel 3 are both reachable backward, and may be used as the final target traffic space.
From the above process, it can be seen that the final goal of the traffic space determination under the Frenet coordinate system based on the reference line is to determine the range of the transverse coordinates that the own vehicle can travel on each longitudinal coordinate s
). Finally, a range of lateral coordinates is continued by a series of longitudinal coordinates s
The final output target traffic space is formed.
Step S208: and planning the track of the automatic driving equipment according to the target passing space.
After the target passing space is determined, the track planning can be performed on the automatic driving equipment. In one possible way, the trajectory planning may be implemented in a conventional manner.
In order to ensure the stability, rationality, anti-interference performance and higher passing efficiency of the track planning, optionally, the target passing space can be closed-loop adjusted according to the previous track planning result; performing space smoothing treatment on the target passing space after closed loop adjustment; according to the result of the space smoothing processing, evaluating the target passing space to obtain a final target passing space; and planning the track of the automatic driving equipment according to the final target passing space.
Specifically, considering that the data of perceived output is unstable in actual engineering practice, there is often a severe jump (including the jump of the existence/non-existence of an obstacle and the jump of the geometric boundary of the obstacle), so as to avoid the occurrence of the severe jump in the final determined output target traffic space, improve the stability of traffic space decision, and also consider the historical output traffic space information and perform necessary fusion to ensure the decision stability, resist the disturbance of perceived data and improve the traffic efficiency. For example, the target traffic space output in the last traffic space decision period can be comprehensively considered on the basis of the target traffic space obtained at the current time point, and the whole space fusion processing is performed by combining the current state of the own vehicle. Under different scenes, specific fusion processing strategies have different strategies and implementation modes according to different engineering requirements, and the strategies and implementation modes can be realized by referring to related technologies, so that the embodiment of the application is not limited.
In addition, due to the light weight/real-time requirement of the scheme provided by the embodiment of the application, in order to ensure that accurate and reasonable results can be output in a downstream scene, the track planning result can be closed loop, and if the downstream planner does not plan a normal result, the running space is correspondingly adjusted, and the traffic space is properly enlarged. Specifically, a closed loop feedback adjustment mode can be adopted to improve the accuracy of the target traffic space decision output. If so, dynamically judging the rationality of the current output target traffic space according to the track information finally planned by the previous track planning, and dynamically adjusting. For example, if a reasonable track is not planned in the last track planning, the range of the target passing space can be properly enlarged, the difficulty of track planning is reduced, and the passing efficiency of the vehicle is improved.
Further, after closed-loop adjustment, space smoothing treatment can be performed on the target traffic space, a part of a unreasonable coordinate range is eliminated, and the quality of track planning is improved, so that smoothness and reasonability of a track generated based on the target traffic space are ensured, and high-efficiency running of the vehicle is ensured. The space smoothing processing mode can be combined with actual engineering requirements, and in the embodiment of the application, optionally, the curvature of the traffic space change can be reduced as much as possible to perform smoothing processing, so that the smoothness of the finally planned track is improved, and the running efficiency of the vehicle is finally improved.
For example, referring to fig. 10, a result of smoothing the traffic space corresponding to the aforementioned passage 2 and the traffic space corresponding to the passage 3 is shown. As can be seen from the figure, when the desired end position is on the left side of the flower bed, the traffic space is concentrated on the left side of the road by smoothing. When the expected end position is on the right side of the flower bed, the traffic space is smooth and reasonable from left to right through smoothing treatment.
After the smoothing process is performed on the target traffic space, the smoothed target traffic space may also be evaluated, for example, the target traffic space may be evaluated according to at least one of the following indexes according to the result of the spatial smoothing process: the method comprises the steps of determining the degree of penetration of automatic driving equipment into a motor vehicle lane, the passing efficiency of the automatic driving equipment, the safety risk of the automatic driving equipment and the matching degree of the state of the automatic driving equipment and a target passing space; and obtaining a final target passing space according to the evaluation result.
The target traffic space (driving channel) obtained through the space smoothing process may have a plurality of target traffic spaces, and an optimal target traffic space is required to be selected from the target traffic spaces as a final result to be output. These target traffic spaces may be evaluated, for example, by way of a cost function, the evaluation term comprising at least one of the following:
(1) Degree of departure from/penetration into a non-motor vehicle lane
For example, small automatic driving devices for logistics distribution generally have low driving speeds and should travel on non-motor lanes to avoid going deep into the motor lanes. So punishment should be made for traffic spaces that deviate from or go deep into a non-motor vehicle lane, then the loss function may employ:
wherein%
, 
) Representing a transverse coordinate range corresponding to a passing space of the automatic driving equipment at a longitudinal coordinate under a Frenet coordinate system, wherein the transverse coordinate range corresponds to the passing space of the automatic driving equipment at the longitudinal coordinate
, 
) Representing a lateral coordinate range corresponding to a non-motor vehicle lane of the autopilot device at a longitudinal coordinate, C 1 Representing losses.
(2) Efficiency of traffic
The passing efficiency directly influences functions exerted by the automatic driving equipment, such as logistics distribution efficiency and user experience, and punishment is carried out on passing space with low passing efficiency, so that a loss function can be adopted:
wherein, the liquid crystal display device comprises a liquid crystal display device,
representing the autopilot device at longitudinal coordinates s in the Frenet coordinate system i The risk of collision of the lateral coordinate range at that location with surrounding obstacles,
representing the longitudinal coordinates s of the autopilot unit i The lateral rate of change of the lateral coordinate range at (like the curvature of a curve),
and (3) with 
Respectively represent
And (3) with
Weight coefficient of C 2 Representing losses.
(3) Safety risk
The safety risk is an evaluation of the potential safety risk of the self-vehicle running in the corresponding traffic space when other surrounding dynamic obstacles (such as dangerous obstacles of high-speed motor vehicles, fast reverse vehicles and the like) are considered. If the safety of the vehicle is affected, corresponding punishment is needed, the punishment can be realized by a person skilled in the art by adopting a proper loss function according to actual requirements, and C is temporarily used 3 And (3) representing. In the embodiment of the present application, the specific implementation manner of the loss function is not limited.
(4) Consistency with degree of match/with historical decision output from current state of vehicle
Switching new traffic space such as not matching with current state (such as speed, direction guiding, etc.) of the own vehicle, if the own vehicle is required to slow down and stop, even reverse, etc., the traffic efficiency of the own vehicle will be affected, and corresponding punishment is required, which can be realized by those skilled in the art according to actual requirements by adopting proper loss function, and C is temporarily used 4 And (3) representing. In the embodiment of the present application, the specific implementation manner of the loss function is not limited.
The calculation results of the above items are weighted and summed according to different weights k, so that the total loss C of each final target traffic space can be obtained, and the total loss C can be expressed as:
C = k 1 * C 1 + k 2 * C 2 + k 3 * C 3 + k 4 * C 4
And selecting a target passing space corresponding to the minimum value from all the total loss C, namely the final target passing space.
On the basis, the track planning of the automatic driving equipment can be performed, including path planning and speed planning. For example, the path planning may plan one or more paths of travel, i.e. curves composed of coordinate location points expected to be reached from the vehicle, based on the resulting target traffic space. The speed planning can be based on the path (curve) obtained by path planning, and the speed/time level planning, namely the time when the self-vehicle is expected to reach the coordinate position, the speed of the self-vehicle, the acceleration and the like can be carried out. And finally obtaining data composed of coordinate position points, time, speed, acceleration information and the like, namely, finally planning a track for the automatic driving equipment.
It can be seen that, according to the embodiment, when determining a traffic space for an autopilot device, a first traffic space that can be driven is determined by a static obstacle removal mode based on map information; then, based on the perception information acquired by the automatic driving equipment, excluding the non-traffic spaces indicated by the perception information from the first traffic space, so as to obtain a second traffic space; and then determining the target traffic space from the second traffic space. On one hand, the process does not need to rely on a complex machine learning model to carry out space decision processing, so that the algorithm complexity is greatly reduced; on the other hand, the trajectory planning of the automatic driving device is performed in the determined target traffic space, namely, the unreasonable traffic space is effectively eliminated, and the data calculation amount of the space decision is further reduced. Therefore, the track planning scheme of the embodiment of the application can be suitable for a vehicle-mounted computing platform with limited calculation force, and a safe and efficient running track can be planned while the light weight of an algorithm is ensured.
The above procedure has been described for the main part of the trajectory planning, based on which an efficient trajectory planning can be achieved. However, in practical applications, in addition to the above-described process, further processing may be performed before and after the above-described process, so as to make the planning more in line with the practical requirements. In the following, the overall process of trajectory planning is illustrated by way of example, as shown in fig. 11.
In fig. 11, the overall process of trajectory planning includes: basic data input, planning preprocessing, traffic space decision, trajectory planning, post decision and output.
Wherein:
the basic data input section mainly inputs: (1) high-precision map data, comprising: the related information of road foundation elements such as roads, lanes, lane lines, road edges, railings/isolation belts and the like, such as position coordinates, topological relations, altitude information and the like; (2) The vehicle positioning data mainly comprises the current position information of the vehicle, such as coordinates, orientation angles and the like; (3) context aware data comprising: the information of the surrounding environment in which the own vehicle is located includes obstacle (dynamic/static) information (geometric information, speed information, type information, etc.), traffic signals, and the like.
The planning preprocessing part performs corresponding preprocessing based on the input basic data, and comprises the following steps: (1) reference line generation, comprising: and dynamically generating a smooth curve capable of guiding the self-vehicle to travel according to the high-precision map data, the self-vehicle positioning data and the coordinate information of the starting point and the ending point of the track planning task, and carrying out coordinate change/projection between a Frenet coordinate system and a Cartesian coordinate system based on the reference line. (2) scene recognition, comprising: a congestion scenario is identified as well as an obstacle (e.g., car/person, etc.) in a congestion waiting state. The obstacle in the congestion waiting state is different from the common static obstacle (i.e. a fixed traffic obstacle), but is only temporarily in the static waiting state for a short time, and often has the characteristics of ordered queuing and the like, and can be regarded as a dynamic obstacle, which is also a factor to be considered in track planning. (3) a prediction process comprising: and predicting the motion trail of the dynamic obstacle around the vehicle in a future period.
On the basis of the planning preprocessing, traffic space decisions and trajectory planning can be performed, which can be implemented by the solution described in the foregoing embodiments, and will not be described in detail here.
The post-decision and output section performs overall evaluation (track evaluation) mainly on the planned plurality of tracks to pick out a track (track selection) meeting the condition therefrom and output, such as the track with the highest evaluation. The overall evaluation method of this section can be implemented in an appropriate manner by those skilled in the art according to actual needs, and the embodiment of the present application is not limited thereto.
Therefore, through the example, a light track planning scheme is provided, the area of the traffic space corresponding to track planning is greatly reduced through the decision of the traffic space, and the number of obstacles required to make the decision of the traffic space is also greatly reduced, so that the calculated amount of a track planning algorithm (optimization/search) is reduced. Meanwhile, due to multiple selection and processing (such as forward and backward searching, smoothing processing and the like) of the traffic space, the shape of the path can be effectively restrained, and unreasonable tracks (such as dangerous and exaggerated paths) can be prevented from being planned, so that safe, reasonable and efficient automatic driving functions such as urban automatic logistics distribution and the like can be realized on a low-calculation-force computing platform.
Referring to fig. 12, a schematic structural view of an automatic driving apparatus according to an embodiment of the present application is shown.
The automatic driving apparatus of the present embodiment includes: a sensor 302, a memory and a processor 304.
The sensor 302 includes, but is not limited to, an image sensor such as a camera, a laser radar sensor, etc., and the sensor 302 may be used to collect environmental data of an environment where the autopilot device is located, such as two-dimensional image data, three-dimensional point cloud data, etc., and generate perception information.
The memory is configured to store at least one executable instruction that causes the processor 304 to perform the trajectory planning process based on the perception information. The specific implementation of the track planning process may be implemented with reference to the related description in the foregoing embodiments, which is not described herein. The processor 404 may be implemented as an on-board embedded chip, for example.
Optionally, the processor 304 may further drive the autopilot device to perform a driving task process according to the trajectory planning result.
Typically, the autopilot unit will also include a drive means 306 and a control means 308. Then, after the processor 304 determines the corresponding trajectory planning result, a driving signal is sent to the driving device 306, and the driving device 306 controls the control device 308 (such as a brake, a throttle, a steering wheel, etc.) in the automatic driving apparatus to perform a corresponding control operation on the automatic driving setting, thereby implementing the autonomous driving of the automatic driving apparatus.
Through the embodiment, the automatic driving equipment can realize safe, reasonable and efficient automatic driving on the low-calculation-force vehicle-mounted computing platform.
The present application further provides a computer storage medium having stored thereon a computer program which, when executed by a processor, implements a trajectory planning method as described in any one of the above-described method embodiments.
Embodiments of the present application also provide a computer program product including computer instructions that instruct a computing device to perform operations corresponding to any one of the track planning methods in the method embodiments described above.
It should be noted that, according to implementation requirements, each component/step described in the embodiments of the present application may be split into more components/steps, and two or more components/steps or part of operations of the components/steps may be combined into new components/steps, so as to achieve the purposes of the embodiments of the present application.
The above-described methods according to embodiments of the present application may be implemented in hardware, firmware, or as software or computer code storable in a recording medium such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk, or as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium and to be stored in a local recording medium downloaded through a network, so that the methods described herein may be stored on such software processes on a recording medium using a general purpose computer, special purpose processor, or programmable or special purpose hardware such as an ASIC or FPGA. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code that, when accessed and executed by a computer, processor, or hardware, performs the methods described herein. Furthermore, when a general purpose computer accesses code for implementing the methods illustrated herein, execution of the code converts the general purpose computer into a special purpose computer for performing the methods illustrated herein.
Those of ordinary skill in the art will appreciate that the elements and method steps of the examples described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or as a combination of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the embodiments of the present application.
The above embodiments are only for illustrating the embodiments of the present application, but not for limiting the embodiments of the present application, and various changes and modifications can be made by one skilled in the relevant art without departing from the spirit and scope of the embodiments of the present application, so that all equivalent technical solutions also fall within the scope of the embodiments of the present application, and the scope of the embodiments of the present application should be defined by the claims.
Claims (12)
1. A trajectory planning method, comprising:
determining a first running space which can be driven for the automatic driving equipment according to map information corresponding to the environment where the automatic driving equipment is located;
According to the perception information collected by the automatic driving device, determining a second passing space for the automatic driving device from the first passing space comprises the following steps: determining a current reference line corresponding to the automatic driving equipment; determining the relation between the transverse displacement and the longitudinal displacement of the automatic driving equipment under the Frenet coordinate system according to the current position of the automatic driving equipment and the curvature of the current reference line; determining a kinematic estimation model matched with the curvature of the current reference line according to the relation; determining a second traffic space for the automatic driving device from the first traffic space according to the determined kinematic estimation model and the perception information acquired by the automatic driving device;
determining a target passing space from the second passing space;
and planning the track of the automatic driving equipment according to the target passing space.
2. The method of claim 1, wherein,
the determining a second traffic space for the autopilot device from the first traffic space according to the determined kinematic estimation model and the perception information collected by the autopilot device comprises: according to the determined kinematic estimation model and the perception information acquired by the automatic driving equipment, forward searching is carried out in the first passing space along the advancing direction of the automatic driving equipment, and at least one second passing space is obtained;
The determining the target traffic space from the second traffic space includes: performing space clipping on the at least one second through space according to the expected end position of the automatic driving equipment; and obtaining a target passing space according to the cutting result.
3. The method of claim 2, wherein the obtaining the target traffic space based on the clipping results comprises:
obtaining at least one candidate passing space according to the clipping result;
performing backward search along the backward direction of the automatic driving equipment in the at least one candidate traffic space according to the determined kinematic estimation model and the perception information acquired by the automatic driving equipment;
and obtaining a target passing space according to the backward search result.
4. The method of claim 1, wherein the determining a relationship of lateral displacement to longitudinal displacement of the autopilot device in a Frenet coordinate system based on the current position of the autopilot device and the curvature of the current reference line comprises:
determining a current reference line of the automatic driving equipment according to the current position of the automatic driving equipment;
judging whether the curvature of the current reference line is 0;
If the curvature of the current reference line is 0, determining that the relation between the transverse displacement and the longitudinal displacement of the automatic driving equipment under the Frenet coordinate system is: the lateral displacement is less than or equal to a product of the longitudinal displacement and a spatial expansion coefficient of the autopilot device in a cartesian coordinate system.
5. The method of claim 4, wherein the method further comprises:
and if the curvature of the current reference line is not 0, determining the relation between the transverse displacement and the longitudinal displacement of the automatic driving equipment under the Frenet coordinate system according to the sine value and the cosine value of the product of the longitudinal displacement and the curvature.
6. A method according to any one of claims 1-3, wherein said trajectory planning of said autopilot device in accordance with said target traffic space comprises:
performing closed-loop adjustment on the target passing space according to the previous track planning result;
performing space smoothing treatment on the target passing space after closed loop adjustment;
according to the result of the space smoothing processing, evaluating the target passing space to obtain a final target passing space;
and planning the track of the automatic driving equipment according to the final target passing space.
7. The method of claim 6, wherein the evaluating the target traffic space based on the result of the spatial smoothing process to obtain a final target traffic space comprises:
based on the result of the spatial smoothing process, the target traffic space is evaluated according to at least one of the following indexes: the degree to which the automatic driving device goes deep into a motor vehicle lane, the passing efficiency of the automatic driving device, the safety risk of the automatic driving device, and the matching degree of the state of the automatic driving device and the target passing space;
and obtaining a final target passing space according to the evaluation result.
8. A method according to any one of claims 1-3, wherein the determining a first available traffic space for the autopilot device according to map information corresponding to an environment in which the autopilot device is located includes:
according to map information corresponding to an environment where automatic driving equipment is located, determining fixed passing barriers in the environment;
and removing the fixed traffic barrier, and determining a first traffic space which can be driven for the automatic driving equipment according to the removed map information.
9. The method of claim 8, wherein the determining a second traffic space for the autopilot device from the first traffic space based on the awareness information collected by the autopilot device comprises:
According to the sensing information acquired by the automatic driving equipment, determining temporary traffic barriers in the environment or non-drivable traffic spaces with the width smaller than a preset width threshold;
excluding the temporary traffic obstacle or the non-travelable traffic space from the first traffic space;
and determining a second passing space for the automatic driving equipment according to the elimination result.
10. An autopilot apparatus comprising: a sensor, a memory, and a processor;
wherein:
the sensor is used for collecting environmental data of the environment where the automatic driving equipment is located and generating perception information;
the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method of any one of claims 1-9 based on the awareness information.
11. A computer storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1-9.
12. A computer program product comprising computer instructions that instruct a computing device to perform operations corresponding to the method of any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310018461.0A CN115892075B (en) | 2023-01-06 | 2023-01-06 | Trajectory planning method, automatic driving device, and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310018461.0A CN115892075B (en) | 2023-01-06 | 2023-01-06 | Trajectory planning method, automatic driving device, and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115892075A CN115892075A (en) | 2023-04-04 |
| CN115892075B true CN115892075B (en) | 2023-06-09 |
Family
ID=86495704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310018461.0A Active CN115892075B (en) | 2023-01-06 | 2023-01-06 | Trajectory planning method, automatic driving device, and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115892075B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6311123B1 (en) * | 1999-06-28 | 2001-10-30 | Hitachi, Ltd. | Vehicle control method and vehicle warning method |
| CN109572695A (en) * | 2018-11-08 | 2019-04-05 | 湖南汽车工程职业学院 | A kind of autonomous driving vehicle Car following control method and system |
| CN113386766A (en) * | 2021-06-17 | 2021-09-14 | 东风汽车集团股份有限公司 | Continuous and periodic self-adaptive synchronous online trajectory planning system and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4389567B2 (en) * | 2003-12-03 | 2009-12-24 | 日産自動車株式会社 | Lane departure prevention device |
| CN109116858B (en) * | 2018-11-07 | 2021-09-07 | 诺亚机器人科技(上海)有限公司 | Obstacle-detouring path planning method and system on designated path |
| US11731612B2 (en) * | 2019-04-30 | 2023-08-22 | Baidu Usa Llc | Neural network approach for parameter learning to speed up planning for complex driving scenarios |
| CN113799797B (en) * | 2021-07-27 | 2022-07-12 | 北京三快在线科技有限公司 | Trajectory planning method and device, storage medium and electronic equipment |
| CN114895677B (en) * | 2022-05-17 | 2023-06-16 | 华南理工大学 | Vehicle track tracking method, system, device and storage medium |
| CN115291597A (en) * | 2022-05-30 | 2022-11-04 | 上海仙途智能科技有限公司 | Path planning method and device based on bidirectional hybrid A-algorithm and terminal |
-
2023
- 2023-01-06 CN CN202310018461.0A patent/CN115892075B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6311123B1 (en) * | 1999-06-28 | 2001-10-30 | Hitachi, Ltd. | Vehicle control method and vehicle warning method |
| CN109572695A (en) * | 2018-11-08 | 2019-04-05 | 湖南汽车工程职业学院 | A kind of autonomous driving vehicle Car following control method and system |
| CN113386766A (en) * | 2021-06-17 | 2021-09-14 | 东风汽车集团股份有限公司 | Continuous and periodic self-adaptive synchronous online trajectory planning system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115892075A (en) | 2023-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11126187B2 (en) | Systems and methods for controlling the operation of a vehicle | |
| CN111123952B (en) | Trajectory planning method and device | |
| Von Hundelshausen et al. | Driving with tentacles: Integral structures for sensing and motion | |
| CN114234998A (en) | Unmanned multi-target-point track parallel planning method based on semantic road map | |
| US20080059015A1 (en) | Software architecture for high-speed traversal of prescribed routes | |
| EP2450763A1 (en) | Global position and orientation estimation system for a vehicle in a passageway environment | |
| CN111397622B (en) | Intelligent automobile local path planning method based on improved A-algorithm and Morphin algorithm | |
| US11815891B2 (en) | End dynamics and constraints relaxation algorithm on optimizing an open space trajectory | |
| CN111830979A (en) | Trajectory optimization method and device | |
| US11586209B2 (en) | Differential dynamic programming (DDP) based planning architecture for autonomous driving vehicles | |
| US11584248B2 (en) | Method of parking an autonomous driving vehicle for autonomous charging | |
| US11628858B2 (en) | Hybrid planning system for autonomous vehicles | |
| CN109709944A (en) | A kind of generation method in enter the station method and its path of entering the station of automatic Pilot bus | |
| CN112859845A (en) | Trajectory planning method and device and vehicle | |
| CN116185014A (en) | Intelligent vehicle global optimal track planning method and system based on dynamic planning | |
| CN113703436A (en) | Parallel barrier buffer system for path planning | |
| EP3842315B1 (en) | Autonomous driving vehicle three-point turn | |
| CN113670305A (en) | Parking trajectory generation method and device, computer equipment and storage medium | |
| CN115140096A (en) | Spline curve and polynomial curve-based automatic driving track planning method | |
| CN115077553A (en) | Method, system, automobile, equipment and medium for planning track based on grid search | |
| CN114715192A (en) | Decoupled real-time trajectory planning method, device and system for automatic driving vehicle | |
| CN115892075B (en) | Trajectory planning method, automatic driving device, and computer storage medium | |
| US11623662B2 (en) | Planning system for speed limit changes for autonomous vehicles | |
| CN116225026A (en) | Automatic driving vehicle operation method, electronic device, and computer-readable medium | |
| CN112644487B (en) | Automatic driving method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231116 Address after: Room 2801, 28th Floor, Building 9, Zone 4, Wangjing Dongyuan, Chaoyang District, Beijing, 100102 Patentee after: Alibaba Damo Academy (Beijing) Technology Co.,Ltd. Address before: 310023 Room 516, floor 5, building 3, No. 969, Wenyi West Road, Wuchang Street, Yuhang District, Hangzhou City, Zhejiang Province Patentee before: Alibaba Dharma Institute (Hangzhou) Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |