A port-oriented automatic driving vehicle control method and device

Abstract

The application discloses a kind of port-oriented automatic driving vehicle control method and device.Therein, the method includes: obtaining forbidden parking area information and the predicted trajectory of reference vehicle;When determining that there is geometric conflict area between vehicle and reference vehicle, the predicted parking position of reference vehicle is determined, and the probability that the position falls into geometric conflict area is determined;The time overlap probability of reference vehicle and vehicle through geometric conflict area is determined;According to the above two probabilities, the probability that vehicle is blocked and stranded after entering forbidden parking area is determined;According to the probability and forbidden parking area information, determine decision instruction;According to the instruction, control vehicle.The application solves the technical problem that in the related art in the complex multi-participant dynamic environment of port, the traditional forbidden parking area decision method based on hard rule cannot quantify the comprehensive risk of "whether vehicle will be blocked and stranded after entering forbidden parking area by other dynamic obstacles", leading to the difficulty in balancing between safety and operating efficiency for automatic driving vehicle (ADV).

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and more specifically, to an autonomous driving vehicle control method and apparatus for ports. Background Technology

[0002] In the context of autonomous driving in ports, traditional decision-making methods rely on fixed, deterministic rules to avoid no-stopping zones such as intersections, crane operating areas, U-shaped curves, and dynamic loading / unloading areas. These methods struggle to cope with the uncertainties arising from dynamic interactions among multiple stakeholders, such as sudden braking by surrounding vehicles, the temporal fluctuations of cranes, and the complexity of spatiotemporal interactions among multiple vehicles. Such methods are either overly conservative, leading to inefficiency, or lack quantitative risk assessment, posing safety hazards. Especially in multi-participant, highly dynamic environments, they struggle to achieve a dynamic trade-off between safety and traffic efficiency, and the decision-making process lacks interpretability and adjustability.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method and apparatus for controlling autonomous vehicles in ports, which at least addresses the technical problem in the related art that traditional no-stopping zone decision-making methods based on hard rules cannot quantify the comprehensive risk of "whether a vehicle will be blocked and delayed by other dynamic obstacles after entering a no-stopping zone" in the complex dynamic environment of ports with multiple participants, making it difficult to balance the safety and operational efficiency of autonomous vehicles (ADVs).

[0005] According to one aspect of the present invention, a method for controlling an autonomous vehicle in a port is provided, comprising: acquiring no-stopping zone information of a no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle; when it is determined that there is a geometric conflict zone between the autonomous vehicle and the reference vehicles based on the planned path of the autonomous vehicle and the predicted trajectories, determining the predicted stopping position of the reference vehicle under braking conditions, and determining a first probability that the predicted stopping position falls into the geometric conflict zone; determining the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone; determining a second probability that the autonomous vehicle will be blocked and delayed after entering the no-stopping zone according to the first probability and the time overlap probability; determining a decision instruction for the autonomous vehicle according to the second probability and the no-stopping zone information; and controlling the autonomous vehicle according to the decision instruction.

[0006] Optionally, determining the geometric conflict zone between the autonomous vehicle and the reference vehicle based on the planned path and the predicted trajectory of the autonomous vehicle includes: performing point-by-point grid sampling of the planned path and the predicted trajectory in a two-dimensional space to obtain grid sampling results; calculating the shortest distance between the planned path and the predicted trajectory within a time window based on the grid sampling results; and determining that the geometric conflict zone exists between the autonomous vehicle and the reference vehicle when the shortest distance between all sampling pairs in the grid sampling results is less than a preset distance threshold.

[0007] Optionally, obtaining the predicted trajectory of reference vehicles around the autonomous vehicle includes: determining the current speed, current acceleration, and current position of the reference vehicles; calculating the initial predicted trajectory of the reference vehicles based on the current acceleration, current speed, and current position; replacing the current acceleration with the instantaneous maximum deceleration of the reference vehicles when an emergency braking command is detected; and determining the predicted trajectory based on the instantaneous maximum deceleration, current speed, and current position.

[0008] Optionally, determining the predicted stopping position of the reference vehicle under braking conditions includes: determining the instantaneous maximum deceleration of the reference vehicle; and determining the predicted stopping position of the reference vehicle after emergency braking based on the instantaneous maximum deceleration, the current position, and the current speed of the reference vehicle.

[0009] Optionally, determining the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone includes: determining the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and determining the estimated time for the reference vehicle to arrive at the geometric conflict zone; determining the time difference between the estimated time and the passage time; and performing probability density analysis on the time difference to obtain the time overlap probability.

[0010] Optionally, determining a second probability that the autonomous vehicle will be blocked and stranded after entering the no-stopping area based on the first probability and the time overlap probability includes: multiplying the first probability and the time overlap probability to obtain the second probability.

[0011] Optionally, determining the decision instruction for the autonomous vehicle based on the second probability and the no-stopping area information includes: determining the no-stopping area type corresponding to the no-stopping area information; selecting a target risk threshold from a preset threshold table based on the no-stopping area type; comparing the second probability with the target risk threshold to obtain a comparison result; determining the decision instruction as a passage instruction when the comparison result indicates that the second probability is less than the target risk threshold, wherein the passage instruction allows the autonomous vehicle to pass through the no-stopping area; and determining the decision instruction as a parking instruction when the comparison result indicates that the second probability is greater than or equal to the target risk threshold, wherein the parking instruction instructs the autonomous vehicle to stop and wait.

[0012] Optionally, the autonomous vehicle control method further includes: when the decision instruction is the passage instruction, dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of the second probability; and when the decision instruction is the stop instruction, updating the planned path of the autonomous vehicle.

[0013] Optionally, controlling the autonomous vehicle according to the decision instruction includes: when the decision instruction is the passage instruction, controlling the autonomous vehicle to travel along the planned path and dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of the second probability; when the decision instruction is the parking instruction, controlling the autonomous vehicle to stop outside the no-parking zone and controlling the autonomous vehicle to travel based on the updated planned path.

[0014] According to another aspect of the present invention, an autonomous vehicle control device for a port is also provided, comprising: an acquisition unit, configured to acquire no-stopping zone information of a no-stopping zone that the autonomous vehicle is about to enter in the port environment and predicted trajectories of reference vehicles around the autonomous vehicle; a first determination unit, configured to determine the predicted stopping position of the reference vehicle under braking conditions and determine a first probability that the predicted stopping position falls into the geometric conflict zone when a geometric conflict zone is determined between the autonomous vehicle and the reference vehicle based on the planned path of the autonomous vehicle and the predicted trajectory; a second determination unit, configured to determine the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone; a third determination unit, configured to determine a second probability that the autonomous vehicle will be blocked and delayed after entering the no-stopping zone based on the first probability and the time overlap probability; a fourth determination unit, configured to determine a decision instruction for the autonomous vehicle based on the second probability and the no-stopping zone information; and a control unit, configured to control the autonomous vehicle according to the decision instruction.

[0015] Optionally, the first determining unit includes: a sampling module, configured to perform point-by-point grid sampling of the planned path and the predicted trajectory in a two-dimensional space to obtain a grid sampling result; a first calculation module, configured to calculate the shortest distance between the planned path and the predicted trajectory within a time window based on the grid sampling result; and a first determining module, configured to determine that the autonomous vehicle and the reference vehicle have the geometric conflict zone when the shortest distance between all sampling pairs in the grid sampling result is less than a preset distance threshold.

[0016] Optionally, the acquisition unit includes: a second determining module, configured to determine the current speed, current acceleration, and current position of the reference vehicle; a second calculating module, configured to calculate the initial predicted trajectory of the reference vehicle based on the current acceleration, the current speed, and the current position; a replacement module, configured to replace the current acceleration with the instantaneous maximum deceleration of the reference vehicle when an emergency braking command is detected; and a third determining module, configured to determine the predicted trajectory based on the instantaneous maximum deceleration, the current speed, and the current position.

[0017] Optionally, the first determining unit includes: a fourth determining module, used to determine the instantaneous maximum deceleration of the reference vehicle; and a fifth determining module, used to determine the predicted stopping position of the reference vehicle after emergency braking based on the instantaneous maximum deceleration of the reference vehicle, its current position, and its current speed.

[0018] Optionally, the second determining unit includes: a sixth determining module, configured to determine the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and to determine the estimated time for the reference vehicle to arrive at the geometric conflict zone; a seventh determining module, configured to determine the time difference between the estimated time and the passage time; and an analysis module, configured to perform probability density analysis on the time difference to obtain the time overlap probability.

[0019] Optionally, the third determining unit includes a third calculation module, used to multiply the first probability and the time overlap probability to obtain the second probability.

[0020] Optionally, the fourth determining unit includes: a corresponding module, configured to select a target risk threshold from a preset threshold table based on the no-parking zone type corresponding to the no-parking zone information; a selection module, configured to select a target risk threshold from a preset threshold table based on the no-parking zone type; a comparison module, configured to compare the second probability with the target risk threshold to obtain a comparison result; an eighth determining module, configured to determine the decision instruction as a passage instruction when the comparison result indicates that the second probability is less than the target risk threshold, wherein the passage instruction allows the autonomous vehicle to pass through the no-parking zone; and a ninth determining module, configured to determine the decision instruction as a parking instruction when the comparison result indicates that the second probability is greater than or equal to the target risk threshold, wherein the parking instruction instructs the autonomous vehicle to stop and wait.

[0021] Optionally, the autonomous vehicle control device further includes: an adjustment unit, configured to dynamically adjust the current speed of the autonomous vehicle according to the magnitude of the second probability when the decision instruction is the passage instruction; and an update unit, configured to update the planned path of the autonomous vehicle when the decision instruction is the stop instruction.

[0022] Optionally, the control unit includes: a first control submodule, configured to control the autonomous vehicle to travel along the planned path when the decision instruction is the passage instruction, and dynamically adjust the current speed of the autonomous vehicle according to the magnitude of the second probability; and a second control submodule, configured to control the autonomous vehicle to stop outside the no-stopping zone when the decision instruction is the parking instruction, and control the autonomous vehicle to travel based on the updated planned path.

[0023] According to another aspect of the present invention, an autonomous vehicle is also provided, which uses the port-oriented autonomous vehicle control method described in any one of the above embodiments.

[0024] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes the port-oriented automated vehicle control method described in any one of the foregoing embodiments.

[0025] According to another aspect of the present invention, a processor is also provided, the processor being configured to run a program, wherein the program, when running, executes the port-oriented automated vehicle control method described in any of the preceding embodiments.

[0026] According to another aspect of the present invention, a computer program product is also provided, including computer instructions that, when executed by a processor, perform the port-oriented automated vehicle control method described in any one of the above embodiments.

[0027] By applying the above-described technical solution of this application, information on no-parking zones and the predicted trajectory of a reference vehicle are obtained; when a geometric conflict zone is determined between a vehicle and a reference vehicle, the predicted parking position of the reference vehicle is determined, and the probability that this position falls into the geometric conflict zone is determined; the probability of time overlap between the reference vehicle and the vehicle passing through the geometric conflict zone is determined; based on the above two probabilities, the probability that the vehicle will be blocked and stranded after entering the no-parking zone is determined; a decision instruction is determined based on this probability and the no-parking zone information; and the vehicle is controlled according to this instruction. This achieves the use of the "probability of being blocked and stranded after entering the no-parking zone" P(trap) as a unified risk indicator, through... By integrating reference vehicle dynamic trajectory prediction (including stochastic modeling of sudden braking deceleration), path space overlap determination, temporal overlap probability calculation, and scene adaptive threshold control, this technology completes universal and probabilistic risk assessment for various no-stopping scenarios such as intersections, track areas, U-turns, and loading / unloading areas. It solves the technical problem in related technologies where traditional no-stopping zone decision-making methods based on hard rules cannot quantify the comprehensive risk of "whether a vehicle will be blocked and delayed by other dynamic obstacles after entering the no-stopping zone" in the complex dynamic environment of ports with multiple participants. This makes it difficult to balance the safety and operational efficiency of autonomous vehicles (ADVs). Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0029] Figure 1 This is a hardware structure block diagram of a mobile terminal for a port-oriented autonomous vehicle control method according to an embodiment of the present invention.

[0030] Figure 2 This is a flowchart of an automated vehicle control method for ports according to an embodiment of the present invention;

[0031] Figure 3 This is a schematic diagram of an automated driving vehicle control device for ports according to an embodiment of the present invention.

[0032] The above figures include the following reference numerals:

[0033] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0034] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0036] As described in the background section, in the complex, multi-participant dynamic environment of ports, traditional no-stopping zone decision-making methods based on hard rules cannot quantify the comprehensive risk of "whether a vehicle will be blocked and delayed by other dynamic obstacles after entering the no-stopping zone," resulting in a difficulty in balancing the safety and operational efficiency of automated vehicles (ADVs). In embodiments of the present invention, an automated vehicle control method and apparatus for ports, an automated vehicle, a computer-readable storage medium, a processor, and a computer program product are provided.

[0037] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0038] The methods and embodiments provided in this invention can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for an automated vehicle control method for ports, according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0039] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the port-oriented autonomous vehicle control method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0040] Example 1

[0041] According to an embodiment of the present invention, a method embodiment of an automated vehicle control method for ports is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0042] Figure 2 This is a flowchart of an automated driving vehicle control method for ports according to an embodiment of the present invention, such as... Figure 2As shown, the method includes the following steps:

[0043] Step S202: Obtain information about the no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle.

[0044] In this embodiment, "no-stopping area information" refers to the boundary and attribute data of specific areas in the port where vehicles are prohibited from stopping, such as the spatiotemporal coordinates of intersection ranges, rail crane operation areas, U-turn restricted areas, or dynamic loading and unloading areas; "predicted trajectory of reference vehicle" refers to predicting the motion path of the vehicle in the future over a period of time by modeling the current position, speed, and acceleration of vehicles around the autonomous vehicle and combining their dynamic characteristics (such as maximum braking capacity). This trajectory is used to evaluate the potential interaction behavior between the vehicle and the autonomous vehicle.

[0045] This method uses high-precision port maps and real-time sensor data to obtain the spatial boundaries and operating rules of the no-stopping zone that the autonomous vehicle is about to enter (such as the operating time of the rail crane, the intersection range, the physical restrictions of the U-shaped bend, etc.). At the same time, it combines the real-time position, speed and acceleration of the surrounding reference vehicles and uses kinematic models (such as uniform acceleration trajectory prediction) to estimate its future short-term motion path, thus forming the predicted trajectory of the reference vehicle.

[0046] By implementing this control method, accurate and quantifiable environmental and object information is provided for the entire risk decision-making process, avoiding reliance on static rules or fuzzy experience. This enables the system to conduct subsequent analysis based on real traffic conditions rather than idealized assumptions, significantly improving the reliability of perception and the quality of decision-making basis for complex port multi-objective interaction scenarios.

[0047] Step S204: When it is determined that there is a geometric conflict zone between the autonomous vehicle and the reference vehicle based on the planned path and predicted trajectory of the autonomous vehicle, the predicted parking position of the reference vehicle under braking condition is determined, and the first probability that the predicted parking position falls into the geometric conflict zone is determined.

[0048] In this embodiment, the "geometric conflict zone" refers to the area in space where the planned path of the autonomous vehicle and the predicted trajectory of the reference vehicle overlap, that is, the physical range where the paths of the two vehicles intersect or cross; the "predicted parking position" refers to the possible final stopping point calculated based on the initial speed and maximum braking deceleration (modeled as random variables) of the reference vehicle when it brakes suddenly due to an emergency; the "first probability" is the probability that the parking position falls into the geometric conflict zone, reflecting whether the reference vehicle will get stuck in the area that the autonomous vehicle must pass through after the sudden braking.

[0049] In this method, after confirming that the planned path of the autonomous vehicle and the predicted trajectory of the reference vehicle have an overlapping area in space (i.e., a geometric conflict zone), the system calculates the theoretical stopping position after emergency braking based on the initial speed and maximum braking deceleration (modeled as random variables following a normal distribution) for the sudden braking behavior of the reference vehicle, and determines whether the stopping position falls within the geometric conflict zone. Then, the probability that the stopping position falls within the conflict zone is calculated, which is the "first probability".

[0050] By implementing this control method, the dynamic risk of "vehicles possibly stopping in the conflict zone due to emergency braking" is transformed into a calculable probability index for the first time. This eliminates the reliance on the binary judgment of "whether a collision will occur" and enables refined and probabilistic modeling of the risk of no-stopping zones, providing a reliable spatial risk basis for subsequent collaborative assessments in the time dimension.

[0051] Step S206: Determine the time overlap probability between the reference vehicle and the autonomous vehicle when passing through the geometric conflict zone.

[0052] In this embodiment, the "time overlap probability" refers to the possibility that there is an overlap between the time required for the reference vehicle to pass through the geometric conflict zone and the time required for the autonomous vehicle to pass through the same zone, and this overlap is determined by the safety time margin. Definition: If the time difference between the two vehicles is less than [a certain value], then [the definition is missing]. If so, it is considered that there is a risk of time conflict.

[0053] This method calculates the time required for an autonomous vehicle to completely exit a pre-determined geometric conflict zone. And the time required for a reference vehicle to arrive at and fully traverse the conflict zone from its current state. And based on the motion state distribution of both, evaluate and Is the time difference between them less than the preset safety time margin? If the time difference is less than If the two vehicles are considered to have a risk of time overlap in the conflict zone, then the probability of this overlap occurring is calculated, which is called the "time overlap probability".

[0054] It should be noted that, This refers to the time it takes for an autonomous vehicle (ADV) to travel from the starting point of entering the conflict zone to the complete departure from the ending point. This time includes the total travel time of the vehicle within the conflict zone. Similarly... This refers to the time it takes for a reference vehicle (i.e., other vehicles relative to the autonomous vehicle) to pass through the conflict zone. Safety time margin, in its physical sense, is the tolerance for reaction delay and uncertainty. Specifically, it reflects the sufficient time given to the system to react when the vehicle makes a decision, and reduces the risk of collision due to control errors or external factors (such as sensor delays). It can be understood as the width of the time window.

[0055] By implementing this control method, the one-sidedness of relying solely on spatial overlap judgment is compensated for. By introducing time-dimensional collaborative constraints, it is ensured that a high-risk judgment is triggered only when two vehicles meet, effectively reducing the false judgment rate. This enables the system to more accurately identify spatiotemporal coupling risks that may actually cause blockage, improving the accuracy of decision-making and adaptability to real-world scenarios.

[0056] Step S208: Determine the second probability that the autonomous vehicle will be blocked and stranded after entering the no-parking zone based on the first probability and the time overlap probability.

[0057] In this embodiment, the "second probability" is a comprehensive risk indicator derived from the first probability and the time overlap probability. It represents the joint probability that an autonomous vehicle will be unexpectedly blocked and stranded by a reference vehicle once it enters a no-parking zone. Its calculation form is as follows: The multiplication of the two reflects the synergistic effect of spatial and temporal risks.

[0058] In this method, the "first probability" (spatial risk of the reference vehicle stopping in the conflict zone after braking) calculated in step S204 is jointly multiplied with the "temporal overlap probability" (temporal risk of overlapping travel times of the two vehicles in the conflict zone) to obtain the comprehensive probability, i.e., the "second probability," of an autonomous vehicle being stuck due to being blocked by the reference vehicle once it enters the no-stopping zone. .

[0059] By implementing this control method, the dispersed spatial and temporal uncertainties are unified into a single, comparable, and decision-making risk quantification value, thereby making the risk explainable, adjustable, and comparable, and providing a clear, mathematical, and engineering practice-compliant basis for subsequent threshold decisions.

[0060] Step S210: Determine the decision instructions for the autonomous vehicle based on the second probability and the no-stopping zone information.

[0061] In this embodiment, "No-Stop Zone Information" is used to associate the safety level setting of the zone, such as a lower risk threshold ε=0.03 for the main work area, and a relaxed threshold of ε=0.05 for secondary passages; the "Decision Instruction" is two types of control signals output after comparing the second probability with the threshold: if If the signal is "Ignore", the vehicle is allowed to proceed at a low speed; otherwise, it is "Yield", requiring the vehicle to stop and wait.

[0062] In this method, the "second probability" calculated in step S208—that is, the comprehensive risk value of an autonomous vehicle being blocked and stranded after entering a no-stopping area—is compared with the safety threshold ε corresponding to the type of the no-stopping area (e.g., ε=0.03 for main work area and ε=0.05 for secondary passage). If the second probability is less than the threshold, an "Ignore" command is output, allowing the vehicle to continue passing under the premise of controlling the vehicle speed; if it is greater than or equal to the threshold, a "Yield" command is output, requiring the vehicle to stop and wait outside the area.

[0063] By implementing this control method, abstract probabilistic risks are transformed into executable control commands, achieving a closed-loop connection from "risk assessment" to "behavioral decision-making." This avoids inefficiency caused by conservative deadlock and prevents accidents caused by reckless passage, significantly improving the port's autonomous driving system's ability to coordinate and optimize safety and operational efficiency in multi-level, highly dynamic environments.

[0064] Step S212: Control the autonomous vehicle according to the decision instructions.

[0065] In this embodiment, the "decision instruction" is the final control signal output by the system. When "Ignore" is given, the vehicle maintains its original path but may decelerate or make minor adjustments to avoid the collision. When "Yield" is given, the vehicle comes to a complete stop and may replan its path to avoid the conflict.

[0066] In this method, based on the decision instruction (Ignore or Yield) output in step S210, the underlying control module of the autonomous vehicle is invoked to execute the corresponding behavior. When the instruction is "Ignore", the vehicle maintains the original planned path, but dynamically reduces speed or fine-tunes the trajectory to reduce the risk margin, while continuously monitoring environmental changes; when the instruction is "Yield", the vehicle safely stops before the boundary of the no-stopping zone, triggering the local replanning module, and regenerating the passage strategy after the risk is eliminated.

[0067] By implementing this control method, a seamless connection is achieved from high-order probabilistic risk decision-making to precise physical layer control. This enables the system not only to determine whether a vehicle should enter a certain area, but also to control how the vehicle enters or stops safely, significantly improving the real-time response, execution reliability, and operational consistency of autonomous vehicles in port scenarios.

[0068] As described above, in this embodiment, the information of the no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle are obtained; when it is determined that there is a geometric conflict zone between the autonomous vehicle and the reference vehicle based on the planned path and predicted trajectory of the autonomous vehicle, the predicted stopping position of the reference vehicle under braking is determined, and the first probability that the predicted stopping position falls into the geometric conflict zone is determined; the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone is determined; the second probability that the autonomous vehicle will be blocked and stuck after entering the no-stopping zone is determined according to the first probability and the time overlap probability; according to the first probability, the second probability that the autonomous vehicle will be blocked and stuck after entering the no-stopping zone is determined; and the third probability that the autonomous vehicle will be blocked and stuck in the port environment is determined according to the second probability. The decision-making instructions for autonomous vehicles are determined by probability and no-stopping zone information. Based on the decision-making instructions, the autonomous vehicles are controlled. By constructing a "joint probability of being blocked after entering" as a unified risk indicator, and integrating spatial overlap analysis, dynamic braking modeling, and spatiotemporal conflict assessment, the decision-making for passage in no-stopping zones is made probabilistic, interpretable, and threshold adjustable. This achieves an effective balance between the safety and passage efficiency of port autonomous vehicles in various types of no-stopping zones, significantly reduces the risk of blockage, collision, and deadlock caused by mistakenly entering no-stopping zones, and supports dynamic adjustment of risk thresholds according to the operation level, achieving the technical effect of intelligent and interpretable real-time decision-making.

[0069] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem in the related art that the traditional no-stopping zone decision-making method based on hard rules cannot quantify the comprehensive risk of "whether a vehicle will be blocked and delayed by other dynamic obstacles after entering the no-stopping zone" in the complex dynamic environment of ports with multiple participants, which makes it difficult to balance the safety and operating efficiency of autonomous vehicles (ADVs).

[0070] According to the above embodiments of the present invention, determining the geometric conflict zone between an autonomous vehicle and a reference vehicle based on the planned path and predicted trajectory of the autonomous vehicle includes: performing point-by-point grid sampling of the planned path and predicted trajectory in a two-dimensional space to obtain grid sampling results; calculating the shortest distance between the planned path and predicted trajectory within a time window based on the grid sampling results; and determining the geometric conflict zone between the autonomous vehicle and the reference vehicle when the shortest distance between all sampling pairs in the grid sampling results is less than a preset distance threshold.

[0071] In this embodiment, the "planned path" refers to the driving trajectory that the autonomous vehicle (ADV) intends to execute within the current decision-making cycle. It is generated by the global path planner and corrected in real time by the dynamic obstacle avoidance module, reflecting the route it intends to travel. The "predicted trajectory" refers to the possible movement path that the reference vehicle (i.e., other non-autonomous vehicles) is extrapolated in a short future window based on its current state (position, speed, acceleration). This trajectory takes into account random behaviors such as sudden braking and is a probabilistic estimate based on the physical model.

[0072] In this method, to accurately determine whether there is spatial overlap between the planned path of the autonomous vehicle (ADV) and the predicted trajectory of the reference vehicle, the system performs grid sampling of the two continuous paths in a two-dimensional plane at a fixed resolution to generate a discrete point set sequence. Then, the Euclidean distance is calculated for all sampling point pairs (ADV points and reference vehicle points), and the minimum value is taken as the shortest spatial interval between the two paths within the time window. Only when the shortest distance of all sampling point pairs is lower than a preset threshold (such as 0.5 meters) is it determined that there is a substantial geometric conflict zone between the two.

[0073] By implementing this control method, the accuracy and noise resistance of collision detection are significantly improved, making it particularly suitable for dynamic environments with curved paths and frequent speed changes in complex port scenarios. At the same time, this method is compatible with any trajectory form (straight line, curve, sampling trajectory), providing a stable and reliable spatial collision basis for subsequent probabilistic risk modeling.

[0074] According to the above embodiments of the present invention, obtaining the predicted trajectory of a reference vehicle around an autonomous vehicle includes: determining the current speed, current acceleration, and current position of the reference vehicle; calculating the initial predicted trajectory of the reference vehicle based on the current acceleration, current speed, and current position; replacing the current acceleration with the instantaneous maximum deceleration of the reference vehicle when an emergency braking command is detected; and determining the predicted trajectory based on the instantaneous maximum deceleration, current speed, and current position.

[0075] In this embodiment, in an intersection scenario, when the ADV is about to enter the intersection area, the system first obtains the motion state vector of the reference vehicle. ,in , , These represent the current position, velocity, and acceleration of the reference vehicle. Assume the reference vehicle will follow the predicted trajectory in the future. The position of the reference vehicle within the short-term prediction window can be represented as: .

[0076] It should be noted that the formula S(t) assumes that the acceleration is a constant value. This is a fundamental assumption used to simplify trajectory prediction calculations. However, this assumption may not be entirely accurate in sudden braking scenarios. During actual braking, the vehicle's acceleration changes, especially during emergency braking, when the vehicle may rapidly reach its maximum braking acceleration. Therefore, in practical applications, when the system detects a sudden braking event, the predicted trajectory should be based on the maximum braking acceleration of the sudden event. Adjustments were made to more accurately reflect the actual braking process. Therefore, the original acceleration... This is only a preliminary prediction; in the event of sudden braking, the instantaneous maximum deceleration should be used. To update the forecast.

[0077] In this method, the system first acquires the current state (position, speed, acceleration) of the reference vehicle in real time based on sensor data, and generates its initial predicted trajectory using a kinematic model (such as a uniform acceleration model). When the system detects that the reference vehicle has triggered an emergency braking command (such as ABS activation, brake light signal, or sudden change in radar echo), the system immediately switches the prediction logic, replaces the original acceleration with the instantaneous maximum deceleration under the vehicle's physical limits, and recalculates the braking termination position and trajectory in combination with the current speed and position, thereby capturing the nonlinear and random characteristics of the high-risk behavior of "sudden braking" in real traffic.

[0078] By implementing this control method, the limitations of traditional linear prediction based on constant acceleration are overcome. For the first time, a dynamic trajectory update mechanism driven by "sudden braking - maximum deceleration" is introduced into the decision-making of autonomous driving in ports, which significantly enhances the prediction accuracy of non-cooperative behavior of reference vehicles. By modeling the maximum deceleration as a probability distribution, not only is the realism of trajectory prediction improved, but also a physically realistic input basis is provided for the subsequent probability calculation of "parking position falling into the conflict zone".

[0079] According to the above embodiments of the present invention, determining the predicted stopping position of a reference vehicle under braking conditions includes: determining the instantaneous maximum deceleration of the reference vehicle; and determining the predicted stopping position of the reference vehicle after emergency braking based on the instantaneous maximum deceleration, the current position, and the current speed of the reference vehicle.

[0080] In this embodiment, considering the possibility of sudden braking by the reference vehicle, its maximum braking deceleration is modeled as a random variable. The corresponding possible parking locations are approximately: .

[0081] It should be noted that, This represents the instantaneous maximum deceleration. This refers to the maximum braking deceleration a vehicle can achieve during emergency braking. Because the vehicle's deceleration during emergency braking is usually instantaneous, not an average value, the formula above should use the instantaneous maximum deceleration.

[0082] In this method, after detecting emergency braking of the reference vehicle, the system no longer relies on its original acceleration, but instead uses its instantaneous maximum deceleration under its physical limits. (Modeled as a random variable following a normal distribution), combined with the current speed and location The theoretical stopping position is calculated based on the kinematic formula of uniform deceleration; this formula implicitly assumes that the braking process is an ideal linear deceleration, but by introducing... The probability distribution characteristics can cover the uncertainty of parking distance for vehicles with different braking capabilities (such as fully loaded / empty, slippery road surface, etc.), thereby constructing a set of predicted parking positions with probabilistic characteristics, rather than a single deterministic value.

[0083] By implementing this control method, the individual differences in vehicle braking performance and environmental uncertainties are quantified into probability distributions for the first time, upgrading parking position prediction from a "deterministic point" to a "risk probability region," providing a scientific and calculable input for subsequent calculations of "whether the parking space falls into the conflict zone." This method significantly improves the system's perception reliability for sudden braking scenarios, avoiding collisions caused by over-avoidance due to conservative estimation or collisions caused by aggressive estimation.

[0084] According to the above embodiments of the present invention, determining the time overlap probability of a reference vehicle and an autonomous vehicle passing through a geometric conflict zone includes: determining the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and determining the estimated time for the reference vehicle to arrive at the geometric conflict zone; determining the time difference between the estimated time and the passage time; and performing probability density analysis on the time difference to obtain the time overlap probability.

[0085] In this embodiment, it is assumed that the predicted trajectory of the reference vehicle is ADV planning path is The area where the two spaces overlap is defined as the conflict zone. .like If there is no geometric conflict, the probability of being trapped is directly zero. Only when spatial overlap exists is the probability of the reference vehicle falling into the conflict zone further calculated. Simultaneously, a time conflict constraint is introduced, assuming the time required for ADV to traverse the conflict zone is... The time it takes for vehicles to arrive at the conflict zone is a random variable. The probability of time overlap is then... ,in For safety time margin.

[0086] Optionally, the time distribution of the preceding vehicle leaving the destination and the arrival time distribution of the ADV are fitted based on historical data. By using historical vehicle operation data, a time distribution model can be fitted, typically using a normal or exponential distribution to describe these time variations. These distribution types can effectively reflect the time fluctuations of vehicles under different operating conditions.

[0087] This method calculates the deterministic passage time required for an autonomous vehicle (ADV) to completely pass through the geometric conflict zone, and, in conjunction with the predicted trajectory of a reference vehicle, estimates its random arrival time at the entrance of the conflict zone. It further calculates the time difference between the two and establishes a probability density function (such as Gaussian distribution or kernel density estimation) based on the uncertainty of the speed and position of the reference vehicle. Finally, it obtains the time overlap probability by performing probability integration on the interval where the time difference is less than the safety margin.

[0088] By implementing this control method, the originally ambiguous question of "whether there will be a time conflict" is transformed into a computable and optimizable probability index, enabling the system not only to identify whether a conflict is possible, but also to assess how likely a conflict is to occur, significantly improving the precision and interpretability of decision-making.

[0089] According to the above embodiments of the present invention, determining the second probability of an autonomous vehicle being blocked and stranded after entering a no-stopping area based on a first probability and a time overlap probability includes: multiplying the first probability and the time overlap probability to obtain the second probability.

[0090] In this embodiment, the overall probability that an ADV will be blocked and stranded after entering a no-stopping zone is defined as follows: .

[0091] This method is based on the independent event joint probability model in probability theory. It regards the "spatial congestion probability" (i.e. the probability that the reference vehicle will fall into the conflict zone after parking) and the "temporal overlap probability" (i.e. the probability that the two vehicles overlap in time in the conflict zone) as two independent key factors affecting the delay of autonomous vehicles (ADVs). The overall risk probability of ADV being blocked and delayed after entering the no-stopping area is calculated by a simple product operation. This product form implies the physical logic that "only when space and time conflict at the same time can delay occur", which is consistent with the reality of the dual constraints of "space occupation + time overlap" in the port scenario.

[0092] By implementing this control method, a simple, interpretable, and adjustable unified risk quantification index was constructed. This transforms the originally complex multidimensional dynamic interaction problem into a single scalar probability value, providing a clear and comparable risk basis for the decision-making system.

[0093] According to the above embodiments of the present invention, determining the decision instruction for an autonomous vehicle based on a second probability and no-stopping zone information includes: determining the no-stopping zone type corresponding to the no-stopping zone information; selecting a target risk threshold from a preset threshold table based on the no-stopping zone type; comparing the second probability with the target risk threshold to obtain a comparison result; determining the decision instruction as a passage instruction when the comparison result indicates that the second probability is less than the target risk threshold, wherein the passage instruction allows the autonomous vehicle to pass through the no-stopping zone; and determining the decision instruction as a parking instruction when the comparison result indicates that the second probability is greater than or equal to the target risk threshold, wherein the parking instruction instructs the autonomous vehicle to stop and wait.

[0094] In this embodiment, the system is configured to ignore risk thresholds. ,when When the condition is met, the output decision is Decision=Ignore (allow passage); otherwise, the output is Yield (wait outside the no-stopping zone). This threshold can be set in stages according to the port's safety level, for example, for the main operating area. The secondary channel can be appropriately widened to This allows for a dynamic balance between security and efficiency.

[0095] In the rail-mounted crane scenario, the spatiotemporal overlap probability of the ADV's (Advanced Driver Assistance Vehicle) stationary track section is calculated by predicting the crane's occupancy range within a time window. If the overlap probability exceeds a set threshold, stopping is prohibited. In the U-shaped restricted turn scenario, the probability of successfully completing the turn is estimated by combining the vehicle's minimum turning radius and the feasible space boundary. If the probability of a failed turn and encroachment on the intersection exceeds a threshold, entry into the turning area is prohibited. In the dynamic loading and unloading operation area scenario, the probability of persistent end-point occupancy is calculated by estimating the time distribution of the preceding vehicle leaving the end point and the arrival time distribution of the ADV. If the occupancy risk exceeds a threshold, the ADV waits behind the preceding vehicle to avoid collisions or deadlocks caused by two vehicles driving side by side.

[0096] It should be noted that the probability of successfully completing the turn is... It is primarily based on a combination of vehicle dynamics simulation and empirical models. The model considers vehicle dynamic constraints (such as maximum steering angle and maximum steering rate) as well as factors such as road friction, steering angle limits, and sideslip. During the simulation, parameters such as the vehicle's sideslip coefficient and road friction coefficient are used to simulate the vehicle's behavior during restricted cornering.

[0097] In this method, the system dynamically matches the corresponding safety threshold from a pre-set graded risk threshold table based on the type of the currently located no-stopping zone (such as intersections, rail crane areas, U-bends, loading and unloading areas, etc.). This threshold table is set differently according to the operational intensity, personnel density, and severity of accident consequences in different areas of the port. Subsequently, the calculated overall congestion probability is... The system compares the data with the target threshold and outputs either a "pass" or a "stop" instruction based on the logic of "pass if the risk is controllable, avoid if the risk exceeds the limit".

[0098] By implementing this control method, "dynamic risk threshold classification" was achieved for the first time in port autonomous driving, enabling the system to strictly control collision risks in high-risk areas (such as next to crane tracks) while maintaining efficient passage in low-risk areas (such as open passages), significantly improving the synergistic optimization of operational efficiency and safety.

[0099] According to the above embodiments of the present invention, the autonomous vehicle control method further includes: when the decision instruction is a passage instruction, dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of the second probability; and when the decision instruction is a stop instruction, updating the planned path of the autonomous vehicle.

[0100] In this embodiment, when the decision is "Ignore," the autonomous vehicle continues along the original path but adjusts its speed (e.g., speed limit or brief deceleration) based on the current risk situation to ensure safety. Path replanning is not involved, but the vehicle dynamically avoids obstacles on the path. When the decision is "Yield," the autonomous vehicle comes to a complete stop, waiting for obstacles or other vehicles to pass. In this case, path replanning may be necessary, especially when waiting for or avoiding complex obstacles. The system will replan based on the current scenario.

[0101] In this method, when the decision instruction is "pass," the system does not simply allow full-speed passage, but rather determines the overall blocking probability based on the calculated probability. The magnitude of the value determines the gradient speed adjustment; the closer the probability is to the threshold, the better. The slower the vehicle slows down, the longer the reaction time and braking distance are allowed. When the decision command is "stop command", the system will actively trigger the path replanning module, which will combine the current obstacle prediction trajectory and the spatiotemporal risk distribution of the no-stopping zone to generate a detour or waiting path to ensure that the vehicle stops in a safe position and avoids secondary conflicts.

[0102] By implementing this control method, the operational flexibility and safety of autonomous vehicles in complex port environments are greatly improved; dynamic speed adjustment avoids the risks of energy consumption and cargo damage caused by sudden braking and acceleration, while path replanning effectively prevents deadlock and path oscillation, significantly enhancing the stability and engineering practicality of the system in multi-vehicle collaboration and highly dynamic scenarios.

[0103] According to the above embodiments of the present invention, controlling an autonomous vehicle according to a decision instruction includes: when the decision instruction is a passage instruction, controlling the autonomous vehicle to travel along a planned path and dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of a second probability; when the decision instruction is a parking instruction, controlling the autonomous vehicle to stop outside a no-parking zone and controlling the autonomous vehicle to travel based on the updated planned path.

[0104] In this embodiment, during path planning, This is the real-time trajectory for dynamic obstacle avoidance. This trajectory relies on global path planning and is adjusted in real-time based on obstacles in the vehicle's surrounding environment. The real-time trajectory is calculated by the dynamic obstacle avoidance module, updating the vehicle's path planning based on sensor data.

[0105] In this method, the system executes differentiated control strategies based on the type of decision instruction. When the instruction is "pass," the vehicle travels along the originally planned route, but the control is adjusted according to the probability of congestion. The vehicle speed is dynamically adjusted based on the probability: the closer the probability is to the threshold, the lower the speed, in order to reserve a safety buffer; when the instruction is "stop", the vehicle immediately brakes at a safe position outside the no-stopping zone boundary and triggers the path replanning module, which combines the current obstacle prediction, spatiotemporal conflict distribution and port operation status to generate a detour or waiting path, ensuring that the vehicle does not intrude into the no-stopping zone and has the ability to pass through efficiently afterward.

[0106] By implementing this control method, abstract probabilistic risk indicators are directly transformed into executable vehicle behavior commands, which not only ensures the physical and operational safety boundaries of the no-stopping zone, but also maintains overall operational efficiency through speed gradient adjustment and path adaptive updates. This mechanism effectively avoids the efficiency loss or safety hazards caused by the traditional extreme strategies of "full speed through" or "absolute prohibition", and significantly improves the coordination and stability of port autonomous vehicles in highly dynamic and multi-constraint environments.

[0107] As can be seen from the above, the technical solution provided by the above embodiments of the present invention uses the "probability of being blocked and stranded after entering a no-stopping area" P (trap) as a unified risk indicator. By integrating reference vehicle dynamic trajectory prediction (including random modeling of sudden braking deceleration), path space overlap determination, time overlap probability calculation, and scene adaptive threshold control, it achieves universal and probabilistic risk assessment for various no-stopping scenarios such as intersections, track areas, U-shaped bends, and loading and unloading areas. This significantly improves the operating efficiency and safety of the port autonomous driving system while ensuring safety.

[0108] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0110] Example 2

[0111] According to embodiments of the present invention, a port-oriented autonomous vehicle control device is also provided for implementing the above-described port-oriented autonomous vehicle control method for a water heater. Figure 3 This is a schematic diagram of an automated driving vehicle control device for ports according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes: an acquisition unit 301, a first determination unit 303, a second determination unit 305, a third determination unit 307, a fourth determination unit 309, and a control unit 311. The device will now be described in detail.

[0112] The acquisition unit 301 is used to acquire information about the no-stopping area that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle.

[0113] The first determining unit 303 is used to determine the predicted parking position of the reference vehicle under braking conditions when a geometric conflict zone is determined between the autonomous vehicle and the reference vehicle based on the planned path and predicted trajectory of the autonomous vehicle, and to determine the first probability that the predicted parking position falls into the geometric conflict zone.

[0114] The second determining unit 305 is used to determine the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone.

[0115] The third determining unit 307 is used to determine the second probability that an autonomous vehicle will be blocked and stranded after entering a no-stopping area based on the first probability and the time overlap probability.

[0116] The fourth determining unit 309 is used to determine the decision instructions for the autonomous vehicle based on the second probability and the no-stopping area information.

[0117] Control unit 311 is used to control the autonomous vehicle according to decision instructions.

[0118] It should be noted that the above-mentioned acquisition unit 301, first determination unit 303, second determination unit 305, third determination unit 307, fourth determination unit 309 and control unit 311 correspond to steps S202 to S212 in the above embodiments. The six units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments.

[0119] As can be seen from the above, in the solution described in the above embodiments of the present invention, an acquisition unit is used to acquire information about the no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectory of reference vehicles around the autonomous vehicle; a first determination unit is used to determine the predicted parking position of the reference vehicle under braking conditions when a geometric conflict zone is determined between the autonomous vehicle and the reference vehicle based on the planned path and predicted trajectory of the autonomous vehicle, and to determine the first probability that the predicted parking position falls into the geometric conflict zone; a second determination unit is used to determine the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone; a third determination unit is used to determine the second probability that the autonomous vehicle will be blocked and stuck after entering the no-stopping zone based on the first probability and the time overlap probability; a fourth determination unit is used to determine the decision command of the autonomous vehicle based on the second probability and the no-stopping zone information; and a control unit is used to control the autonomous vehicle according to the decision command. The above technical solution achieves the goal of probabilistic, interpretable, and threshold-adjustable decision-making for passage in no-stopping areas by constructing a "joint probability of being blocked after entry" as a unified risk indicator, integrating spatial overlap analysis, dynamic braking modeling, and spatiotemporal conflict assessment. This effectively balances the safety and passage efficiency of autonomous vehicles in various types of no-stopping areas in ports, significantly reduces the risk of blockage, collision, and deadlock caused by mistakenly entering no-stopping areas, and supports dynamic adjustment of risk thresholds according to the operation level, achieving the technical effect of intelligent and interpretable real-time decision-making.

[0120] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem in the related art that the traditional no-stopping zone decision-making method based on hard rules cannot quantify the comprehensive risk of "whether a vehicle will be blocked and delayed by other dynamic obstacles after entering the no-stopping zone" in the complex dynamic environment of ports with multiple participants, which makes it difficult to balance the safety and operating efficiency of autonomous vehicles (ADVs).

[0121] Optionally, the first determining unit includes: a sampling module, used to perform point-by-point grid sampling of the planned path and the predicted trajectory in a two-dimensional space to obtain grid sampling results; a first calculation module, used to calculate the shortest distance between the planned path and the predicted trajectory within a time window based on the grid sampling results; and a first determining module, used to determine that there is a geometric conflict zone between the autonomous vehicle and the reference vehicle when the shortest distance between all sampling pairs in the grid sampling results is less than a preset distance threshold.

[0122] Optionally, the acquisition unit includes: a second determining module for determining the current speed, current acceleration, and current position of the reference vehicle; a second calculating module for calculating the initial predicted trajectory of the reference vehicle based on the current acceleration, current speed, and current position; a replacement module for replacing the current acceleration with the instantaneous maximum deceleration of the reference vehicle when an emergency braking command is detected; and a third determining module for determining the predicted trajectory based on the instantaneous maximum deceleration, current speed, and current position.

[0123] Optionally, the first determining unit includes: a fourth determining module for determining the instantaneous maximum deceleration of the reference vehicle; and a fifth determining module for determining the predicted stopping position of the reference vehicle after emergency braking based on the instantaneous maximum deceleration of the reference vehicle, its current position, and its current speed.

[0124] Optionally, the second determining unit includes: a sixth determining module, used to determine the passage time required for the autonomous vehicle to pass through the geometric conflict zone and to determine the estimated time for the reference vehicle to arrive at the geometric conflict zone; a seventh determining module, used to determine the time difference between the estimated time and the passage time; and an analysis module, used to perform probability density analysis on the time difference to obtain the time overlap probability.

[0125] Optionally, the third determining unit includes: a third calculation module, used to multiply the first probability and the time overlap probability to obtain the second probability.

[0126] Optionally, the fourth determining unit includes: a corresponding module, used to determine the type of no-parking area corresponding to the no-parking area information; a selection module, used to select a target risk threshold from a preset threshold table according to the no-parking area type; a comparison module, used to compare the second probability with the target risk threshold to obtain a comparison result; an eighth determining module, used to determine the decision instruction as a passage instruction when the comparison result indicates that the second probability is less than the target risk threshold, wherein the passage instruction allows the autonomous vehicle to pass through the no-parking area; and a ninth determining module, used to determine the decision instruction as a parking instruction when the comparison result indicates that the second probability is greater than or equal to the target risk threshold, wherein the parking instruction instructs the autonomous vehicle to stop and wait.

[0127] Optionally, the autonomous vehicle control device further includes: an adjustment unit, used to dynamically adjust the current speed of the autonomous vehicle according to the magnitude of a second probability when the decision instruction is a passage instruction; and an update unit, used to update the planned path of the autonomous vehicle when the decision instruction is a stop instruction.

[0128] Optionally, the control unit includes: a first control submodule, configured to control the autonomous vehicle to travel along the planned path when the decision instruction is a passage instruction, and dynamically adjust the current speed of the autonomous vehicle according to the magnitude of the second probability; and a second control submodule, configured to control the autonomous vehicle to stop outside the no-stopping zone when the decision instruction is a stop instruction, and control the autonomous vehicle to travel based on the updated planned path.

[0129] According to another aspect of the present invention, an autonomous vehicle is also provided, which uses any of the above-described port-oriented autonomous vehicle control methods.

[0130] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes any of the above-described methods for controlling automated vehicles for ports.

[0131] According to another aspect of the present invention, a processor is also provided, which is used to run a program, wherein the program executes any of the above-described methods for controlling automated vehicles oriented towards ports.

[0132] According to another aspect of the present invention, a computer program product is also provided, including computer instructions that, when executed by a processor, perform any of the above-described methods for controlling automated vehicles for ports.

[0133] Optionally, in this embodiment, the computer-readable storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any communication device in a group of communication devices.

[0134] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: acquiring information about the no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle; when it is determined that there is a geometric conflict zone between the autonomous vehicle and the reference vehicles based on the planned path and predicted trajectory of the autonomous vehicle, determining the predicted stopping position of the reference vehicle under braking conditions, and determining a first probability that the predicted stopping position falls into the geometric conflict zone; determining the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone; determining a second probability that the autonomous vehicle will be blocked and delayed after entering the no-stopping zone based on the first probability and the time overlap probability; determining the decision command of the autonomous vehicle based on the second probability and the no-stopping zone information; and controlling the autonomous vehicle according to the decision command.

[0135] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: performing point-by-point grid sampling of the planned path and the predicted trajectory in a two-dimensional space to obtain grid sampling results; calculating the shortest distance between the planned path and the predicted trajectory within a time window based on the grid sampling results; and determining that there is a geometric conflict zone between the autonomous vehicle and the reference vehicle when the shortest distance between all sampling pairs in the grid sampling results is less than a preset distance threshold.

[0136] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining the current speed, current acceleration, and current position of the reference vehicle; calculating the initial predicted trajectory of the reference vehicle based on the current acceleration, current speed, and current position; replacing the current acceleration with the instantaneous maximum deceleration of the reference vehicle when an emergency braking command is detected; and determining the predicted trajectory based on the instantaneous maximum deceleration, current speed, and current position.

[0137] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining the instantaneous maximum deceleration of the reference vehicle; and determining the predicted stopping position of the reference vehicle after emergency braking based on the instantaneous maximum deceleration of the reference vehicle, the current position, and the current speed.

[0138] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and determining the estimated time for the reference vehicle to arrive at the geometric conflict zone; determining the time difference between the estimated time and the passage time; and performing probability density analysis on the time difference to obtain the time overlap probability.

[0139] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: multiplying the first probability with the time overlap probability to obtain the second probability.

[0140] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: according to the no-stopping area type corresponding to the no-stopping area information; selecting a target risk threshold from a preset threshold table according to the no-stopping area type; comparing a second probability with the target risk threshold to obtain a comparison result; when the comparison result indicates that the second probability is less than the target risk threshold, determining the decision instruction as a passage instruction, wherein the passage instruction allows the autonomous vehicle to pass through the no-stopping area; when the comparison result indicates that the second probability is greater than or equal to the target risk threshold, determining the decision instruction as a parking instruction, wherein the parking instruction instructs the autonomous vehicle to stop and wait.

[0141] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: when the decision instruction is a pass instruction, dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of the second probability; when the decision instruction is a stop instruction, updating the planned path of the autonomous vehicle.

[0142] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: when the decision instruction is a pass instruction, controlling the autonomous vehicle to travel along the planned path and dynamically adjusting the current speed of the autonomous vehicle according to the magnitude of the second probability; when the decision instruction is a stop instruction, controlling the autonomous vehicle to stop outside the no-stopping zone and controlling the autonomous vehicle to travel based on the updated planned path.

[0143] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be 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 displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0145] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0147] Furthermore, the functional units in the various embodiments of the present invention 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0148] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part 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 the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0149] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for controlling automated vehicles in ports, characterized in that, include: Obtain information on the no-stopping zone that the autonomous vehicle is about to enter in the port environment, as well as the predicted trajectories of reference vehicles around the autonomous vehicle. When a geometric conflict zone is determined between the autonomous vehicle and the reference vehicle based on the planned path and the predicted trajectory of the autonomous vehicle, the predicted parking position of the reference vehicle under braking conditions is determined, and a first probability that the predicted parking position falls into the geometric conflict zone is determined. The geometric conflict zone is determined as follows: the planned path and the predicted trajectory are sampled point-by-point in a two-dimensional space to obtain a grid sampling result; the shortest distance between the planned path and the predicted trajectory within a time window is calculated based on the grid sampling result; when the shortest distance between all sampling pairs in the grid sampling result is less than a preset distance threshold, it is determined that the geometric conflict zone exists between the autonomous vehicle and the reference vehicle. The time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone is determined, wherein the time overlap probability is determined by: determining the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and determining the estimated time for the reference vehicle to arrive at the geometric conflict zone; determining the time difference between the estimated time and the passage time; and performing probability density analysis on the time difference to obtain the time overlap probability. A second probability is determined based on the first probability and the time overlap probability, after the autonomous vehicle enters the no-stopping area, it is blocked and stranded. The decision instructions for the autonomous vehicle are determined based on the second probability and the no-parking zone information; The autonomous vehicle is controlled according to the decision instructions.

2. The port-oriented automated vehicle control method according to claim 1, characterized in that, Obtaining the predicted trajectories of reference vehicles around the autonomous vehicle includes: Determine the current speed, current acceleration, and current position of the reference vehicle; The initial predicted trajectory of the reference vehicle is calculated based on the current acceleration, the current velocity, and the current position; When an emergency braking command is detected for the reference vehicle, the current acceleration is replaced with the instantaneous maximum deceleration of the reference vehicle. The predicted trajectory is determined based on the instantaneous maximum deceleration, the current velocity, and the current position.

3. The port-oriented automated vehicle control method according to claim 1, characterized in that, Determining the predicted stopping position of the reference vehicle under braking conditions includes: Determine the instantaneous maximum deceleration of the reference vehicle; The predicted stopping position of the reference vehicle after emergency braking is determined based on the instantaneous maximum deceleration, current position, and current speed of the reference vehicle.

4. The port-oriented automated vehicle control method according to claim 1, characterized in that, Determining a second probability based on the first probability and the time overlap probability of the autonomous vehicle being blocked and stranded after entering the no-stopping zone includes: The second probability is obtained by multiplying the first probability by the time overlap probability.

5. The port-oriented automated vehicle control method according to any one of claims 1 to 4, characterized in that, The decision instructions for the autonomous vehicle are determined based on the second probability and the no-stopping zone information, including: Based on the type of no-parking zone corresponding to the no-parking zone information; Select the target risk threshold from the preset threshold table based on the type of no-parking zone; The second probability is compared with the target risk threshold to obtain the comparison result; When the comparison result indicates that the second probability is less than the target risk threshold, the decision instruction is determined to be a passage instruction, wherein the passage instruction allows the autonomous vehicle to pass through the no-stopping area; When the comparison result indicates that the second probability is greater than or equal to the target risk threshold, the decision instruction is determined to be a parking instruction, wherein the parking instruction is used to instruct the autonomous vehicle to stop and wait.

6. The port-oriented automated vehicle control method according to claim 5, characterized in that, The autonomous vehicle control method further includes: When the decision instruction is the passage instruction, the current speed of the autonomous vehicle is dynamically adjusted according to the magnitude of the second probability; When the decision instruction is the parking instruction, the planned path of the autonomous vehicle is updated.

7. The port-oriented automated vehicle control method according to claim 5, characterized in that, Controlling the autonomous vehicle according to the decision instructions includes: When the decision instruction is the passage instruction, the autonomous vehicle is controlled to travel along the planned path, and the current speed of the autonomous vehicle is dynamically adjusted according to the magnitude of the second probability. When the decision instruction is the parking instruction, the autonomous vehicle is controlled to stop outside the no-parking zone, and the autonomous vehicle is controlled to travel based on the updated planned path.

8. An automated driving vehicle control device for ports, characterized in that, include: The acquisition unit is used to acquire information about the no-stopping zone that the autonomous vehicle is about to enter in the port environment and the predicted trajectories of reference vehicles around the autonomous vehicle. The first determining unit is configured to, when determining that a geometric conflict zone exists between the autonomous vehicle and the reference vehicle based on the planned path and the predicted trajectory of the autonomous vehicle, determine the predicted parking position of the reference vehicle under braking conditions, and determine a first probability that the predicted parking position falls into the geometric conflict zone, wherein the geometric conflict zone is determined based on the following method: performing point-by-point grid sampling on the planned path and the predicted trajectory in a two-dimensional space to obtain grid sampling results; calculating the shortest distance between the planned path and the predicted trajectory within a time window based on the grid sampling results; and determining that the geometric conflict zone exists between the autonomous vehicle and the reference vehicle when the shortest distance between all sampling pairs in the grid sampling results is less than a preset distance threshold. The second determining unit is configured to determine the time overlap probability of the reference vehicle and the autonomous vehicle passing through the geometric conflict zone, wherein the time overlap probability is determined by: determining the passage time required for the autonomous vehicle to pass through the geometric conflict zone, and determining the estimated time for the reference vehicle to arrive at the geometric conflict zone; determining the time difference between the estimated time and the passage time; and performing probability density analysis on the time difference to obtain the time overlap probability. The third determining unit is used to determine a second probability that the autonomous vehicle will be blocked and stranded after entering the no-stopping area based on the first probability and the time overlap probability. The fourth determining unit is used to determine the decision instruction of the autonomous vehicle based on the second probability and the no-stopping area information; A control unit for controlling the autonomous vehicle according to the decision instructions.

Images