Vehicle light signal control methods, vehicle storage media
By acquiring vehicle driving environment information and generating dynamic light signals, the problem of existing vehicle lighting systems being unable to convey driving intentions has been solved, achieving improvements in safety and efficiency in complex traffic scenarios.
Patent Information
- Application Number
- CN202611230048.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-13
- Publication Date
- 2026-11-17
AI Technical Summary
Existing vehicle lighting systems are unable to proactively, on demand, and in a targeted manner convey visual signals related to driving intentions in complex traffic scenarios, making it difficult for other road users to predict vehicle actions and easily causing confusion or misunderstanding.
By acquiring vehicle driving environment information, it determines whether the current traffic scene meets the conditions for expressing driving intentions, generates corresponding intention expression information, and controls the vehicle's lighting device to perform dynamic light signal expression, including adjusting the sweeping frequency and flashing cycle, to achieve light signal control.
It improves the safety and communication efficiency of human-vehicle interaction in mixed traffic environments, reduces interference from invalid light signal output to traffic participants, and enhances social acceptance.
Smart Images

Figure CN123104516A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle control technology, and in particular to a method for controlling vehicle lights, a vehicle, and a storage medium. Background Technology
[0002] With the increasing complexity of road traffic environments, vehicles need to effectively interact with other road users (such as pedestrians, non-motorized vehicles, and other motor vehicles) to ensure safe and smooth driving. Currently, vehicles are typically equipped with standardized external lighting systems, such as turn signals, brake lights, and hazard warning lights. The functions and semantics of these light signals are strictly defined by traffic regulations, indicating basic operational information such as the vehicle's turning intention, deceleration and braking status, or temporary stopping. When the driver operates the corresponding switch, these lights operate according to a fixed illumination pattern (such as constant illumination or uniform flashing), conveying clear and singular operational instructions to the outside world.
[0003] However, in many real-world traffic scenarios, especially in unstructured traffic environments such as intersections without traffic lights, narrow sections of road without clear right-of-way signs, or mixed pedestrian and vehicle traffic, a vehicle's actual driving intentions do not always perfectly correspond to basic operational instructions. For example, a vehicle may slow down to observe road conditions but not intend to stop, or it may temporarily stop to wait based on the movement of oncoming vehicles, but this does not constitute a breakdown stop. In these situations, the vehicle's existing lighting system can only express its physical operating status (such as whether it is braking or steering), but cannot reflect its current driving decision or its next course of action.
[0004] Because existing vehicle lighting signal systems have limited functionality and fixed semantics, they cannot proactively convey intent information related to different traffic scenarios, and their illumination patterns do not change differentiatedly with varying scenarios. This makes it difficult for other road users to anticipate the actions a vehicle with unclear intentions will take, easily leading to confusion or misunderstanding.
[0005] Therefore, how to enable vehicles to actively, on demand, and in a targeted manner transmit visual signals related to the current driving intention in complex traffic scenarios has become an urgent technical problem to be solved. Summary of the Invention
[0006] The purpose of this application is to provide a vehicle light signal control method, a vehicle, and a storage medium that can actively, on demand, and purposefully transmit visual signals related to the current driving intention to the outside world.
[0007] To achieve the above objectives: In a first aspect, a method for controlling vehicle lighting signals includes: acquiring vehicle driving environment information; determining, based on the driving environment information, whether the current traffic scene meets the conditions for expressing driving intention; in response to meeting the conditions for expressing driving intention, generating corresponding intention expression information based on the current traffic scene; determining corresponding first light effect information based on the intention expression information; generating a lighting signal control command based on the first light effect information, and controlling the vehicle's light-emitting device to execute the lighting signal expression based on the lighting signal control command.
[0008] In one embodiment, determining whether the current traffic scenario meets the conditions for expressing driving intent based on driving environment information includes: acquiring road topology information and motion state information of traffic participants from the driving environment information; determining whether the current traffic scenario is a preset scenario based on the road topology information and motion state information; and determining whether the current traffic scenario meets the conditions for expressing driving intent if the current traffic scenario is a preset scenario.
[0009] In one embodiment, determining whether the current traffic scenario is a preset scenario based on road topology information and motion state information includes: detecting whether there is traffic light control or clear right-of-way signs on the current road; if there is no traffic light control and no right-of-way signs, and at least one target traffic participant that conflicts with the path of the vehicle is detected, then the current traffic scenario is determined to be a preset scenario.
[0010] In one embodiment, generating corresponding intent expression information based on the current traffic scenario includes: obtaining a set of candidate driving strategies for the vehicle in the current traffic scenario; determining the current driving intent type to be expressed based on the evaluation benefit value of each candidate strategy in the set of candidate driving strategies; generating a driving intent confidence parameter corresponding to the driving intent type based on the confidence evaluation result of the driving intent type; and combining the driving intent type and the corresponding driving intent confidence parameter to generate intent expression information.
[0011] In one embodiment, the driving intention type to be expressed is determined based on the evaluated payoff value of each candidate strategy in the candidate driving strategy set, including at least one of the following: when the candidate driving strategy set includes a deceleration and waiting strategy and an acceleration and passage strategy and the difference between their evaluated payoff values is less than a preset threshold, the driving intention type is determined to be a hesitant intention; when the driving behavior corresponding to the highest payoff strategy in the candidate driving strategy set is passage and the difference between the highest payoff value and the payoff values of other candidate strategies is less than a second preset threshold, the driving intention type is determined to be a cautious passage intention or a game-theoretic confirmation intention.
[0012] In one embodiment, the first light effect information includes a sweeping frequency and / or a flashing period. Determining the corresponding first light effect information based on the intent expression information includes: extracting the driving intent type and the corresponding driving intent confidence parameter from the intent expression information; using the driving intent type as a query index, searching for a basic light effect pattern that matches the intent type from a preset mapping relationship, wherein the preset mapping relationship includes the correspondence between the driving intent type and the basic light effect pattern; adjusting at least one dynamic parameter in the basic light effect pattern according to the driving intent confidence parameter, and using the adjusted light effect pattern as the first light effect information; wherein adjusting the dynamic parameter includes: controlling the driving intent confidence parameter to have a preset negative correlation with the sweeping frequency and / or flashing period in the first light effect information.
[0013] In one embodiment, generating a light control command based on the first light effect information includes: obtaining a first intention priority corresponding to the first light effect information; obtaining a second intention priority corresponding to the second light effect information currently being executed by the light-emitting device; comparing the first intention priority with the enemy intention priority; if the first intention priority is higher than the second intention priority, generating a light control command corresponding to the first light effect information; if the first intention priority is lower than the second intention priority, generating a light control command corresponding to the second light effect information.
[0014] In one embodiment, the acquisition of intent priority includes: acquiring a first driving intent type corresponding to the first light effect information and a second driving intent type corresponding to the second light effect information; when the first driving intent type or the second driving intent type is a hesitant intent, assigning a waiting decision-making priority; when the first driving intent type or the second driving intent type is a cautious passage intent or a game-theoretic confirmation intent, assigning an efficiency passage priority; and determining the first intent priority and the second intent priority according to a preset priority rule, wherein the waiting decision-making priority is lower than the efficiency passage priority.
[0015] Secondly, embodiments of this application provide a vehicle, specifically including: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions for performing the vehicle light signal control method as described in the first aspect.
[0016] Thirdly, embodiments of this application provide a machine-readable storage medium storing a computer program that, when the instructions in the machine-readable storage medium are executed by a vehicle's processor, enables the vehicle to implement the vehicle's light signal control method as described in the first aspect.
[0017] This application provides a vehicle light signal control method, a vehicle, and a machine-readable storage medium, comprising: acquiring vehicle driving environment information; determining whether the current traffic scene meets the driving intention expression conditions based on the driving environment information; generating corresponding intention expression information based on the current traffic scene in response to meeting the driving intention expression conditions; determining corresponding first light effect information based on the intention expression information; generating light signal control instructions based on the first light effect information, and controlling the vehicle's light-emitting device to execute the light signal expression based on the light signal control instructions. This application, by acquiring vehicle driving environment information and determining whether the current traffic scene meets the driving intention expression conditions based on the driving environment information, can trigger subsequent light signal control processes when the conditions are met, realizing on-demand triggering of light signal expression rather than continuous unconditional execution; furthermore, by generating corresponding intention expression information based on the current traffic scene and determining corresponding first light effect information based on the intention expression information, the expressed light effect is associated with the current traffic scene, enabling the vehicle to execute differentiated light signal expressions in different traffic scenes; further still, by automatically generating light signal control instructions based on the first light effect information and controlling the light-emitting device to execute the light signal expression, the entire light signal expression process is automated without manual intervention. Furthermore, through the above embodiments, the vehicle is able to actively output light signals when certain conditions are met, thereby realizing the active expression of driving intentions. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating a vehicle light signal control method provided in an embodiment of the present invention.
[0019] Figure 2 This is a schematic diagram illustrating the specific process of the vehicle light signal control method provided in an embodiment of the present invention.
[0020] Figure 3 This is a structural schematic diagram of a vehicle provided in an embodiment of the present invention.
[0021] Processor 410, memory 411, network interface 412, bus system 413. Detailed Implementation
[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0023] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.
[0024] It should be understood that although the terms first, second, third, etc., may be used herein to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this document, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if," as used herein, can be interpreted as "when," "when," or "in response to determination." Furthermore, as used herein, the singular forms "a," "an," and "the" are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprising," "including," indicate the presence of the stated feature, step, operation, element, component, item, kind, and / or group, but do not exclude the presence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" as used herein are to be interpreted as inclusive, or mean any one or any combination thereof. Therefore, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A, B, and C". Exceptions to this definition will only occur if the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.
[0025] It should be understood that although the steps in the flowcharts of this application's embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.
[0026] It should be noted that step designations such as S101 and S102 are used in this document for the purpose of more clearly and concisely describing the corresponding content, and do not constitute a substantial limitation on the order. In specific implementation, those skilled in the art may execute S102 first and then S101, etc., but these should all be within the protection scope of this application.
[0027] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.
[0028] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.
[0029] It should be noted that currently, the functional semantics of vehicle external lights (such as turn signals, brake lights, and hazard warning lights) are strictly limited to the transmission of basic operational commands such as steering, braking, and malfunctions. This is a passive information expression with fixed semantics, triggered directly by the driver. While vehicles with advanced autonomous driving capabilities or advanced driver assistance systems (ADAS) can perceive the surrounding traffic environment in real time and generate corresponding driving strategies (such as assessing right-of-way at unsignalized intersections and determining whether to yield when meeting oncoming traffic on narrow roads), this crucial information reflecting the vehicle's driving decision-making tendencies and interactive intentions only serves the vehicle's own operational planning. It is not effectively communicated to other traffic participants (such as drivers of manually operated vehicles, pedestrians, and non-motorized vehicle riders) through visual signals. This results in other traffic participants being unable to understand the vehicle's next action tendency in a timely and accurate manner, which is a significant factor contributing to traffic standstill at intersections, frequent emergency braking, low traffic efficiency, and a lack of trust between humans and vehicles in mixed traffic environments. The existing technology contains information gaps regarding the "opacity of intent" of autonomous vehicles, leading to a waste of perception resources and safety hazards.
[0030] To address the aforementioned issues, this application proposes a vehicle lighting signal control method. This method allows vehicles to utilize their environmental perception and decision-making system to acquire real-time driving environment information and automatically determine whether the current traffic scene requires expressing driving intentions. When the intention expression conditions are met, the vehicle automatically generates corresponding intention expression information based on the current traffic scene. A preset mapping relationship is used to determine the matching lighting effect information. Finally, the method controls a light-emitting device located outside the vehicle and independent of conventional light groups to output light using corresponding dynamic lighting signals. This intuitively conveys the vehicle's driving decision-making status and interaction tendencies (such as weighing options, preparing to proceed cautiously, or actively yielding) to surrounding traffic participants. This fills the technological gap in existing vehicle lighting systems that cannot express advanced social intentions, improving the safety, communication efficiency, and social acceptance of human-vehicle interaction in mixed traffic environments. It can be widely applied to various types of L3 and above autonomous driving vehicles with environmental perception and decision-making capabilities, advanced driver assistance vehicles, and commercial vehicles with intelligent interaction needs.
[0031] like Figure 1 As shown, this application embodiment proposes a vehicle light signal control method. This method can be implemented in software and / or hardware, for example, as a vehicle controller or intelligent driving domain controller integrating an environmental perception and decision-making system. This embodiment uses a vehicle with intelligent driving capabilities as an example. The vehicle can acquire real-time driving environment information through its perception modules (e.g., cameras, lidar, millimeter-wave radar, V2X communication equipment, etc.) and can perform scene recognition and driving intention generation based on the perceived environmental data using a built-in decision-making algorithm. The vehicle light signal control method provided in this application embodiment includes the following steps: Step S101: Obtain vehicle driving environment information.
[0032] Optionally, driving environment information refers to various environmental data related to the current traffic scenario collected by the vehicle's onboard perception system during driving. This information is used to support the vehicle's identification and judgment of the current traffic scenario.
[0033] For example, driving environment information may include: road topology information, traffic participant information, and traffic environment attribute information. Specifically, road topology information refers to the road structure data of the current road segment and surrounding area where the vehicle is located, including but not limited to: lane distribution, intersection shape and type (such as crossroads, T-junctions), distribution of traffic lights and traffic signs, and the presence and location of right-of-way signs (such as stop signs, yield signs), etc. Road topology information may be derived from image data collected by vehicle-mounted cameras, high-precision map data, or road information broadcast by V2X roadside equipment. Traffic participant information refers to the status data of other traffic participants who are related to or potentially conflict with the vehicle's driving path in the current traffic scenario, including but not limited to: the position, speed, acceleration, driving direction, and trajectory of other motor vehicles; the position and movement status of non-motorized vehicles (such as bicycles, electric bicycles); and the position, movement direction, and crossing intention of pedestrians, etc. The above information can be obtained in real time through sensors such as cameras, lidar, and millimeter-wave radar. Traffic environment attribute information refers to the environmental status data of the current traffic scenario, including but not limited to: road traffic status (such as congestion, smooth traffic), weather and lighting conditions (affecting sensor perception commands), traffic light status (red / green / yellow), etc.
[0034] In addition, for example, in vehicles equipped with V2X communication capabilities, driving environment information may also include driving status information broadcast by other vehicles and road event information sent by roadside equipment, which are received via V2X communication.
[0035] Step S102: Based on the driving environment information, determine whether the current traffic scenario meets the conditions for expressing driving intention.
[0036] Optionally, the driving intent expression condition refers to the threshold condition used to determine whether a vehicle needs to output light signals in the current traffic scenario to convey its driving decision tendency to surrounding traffic participants. The existence of this condition is to avoid the vehicle indiscriminately executing light signals in all driving scenarios, thereby ensuring that light signal output is only triggered in specific scenarios where it is truly necessary to communicate driving intent.
[0037] In other words, the essence of the driving intent expression condition is to determine, based on the driving environment information, whether the current traffic situation is one where right-of-way is unclear and requires communication of intent to proceed. Only when this condition is met will the vehicle trigger the subsequent light signal expression process; if this condition is not met, such as at a traffic light-controlled intersection or on an open road with no other road users ahead, the vehicle will not execute the light signal expression to reduce invalid or unnecessary light signal output that may interfere with other road users.
[0038] In some implementations, step S102, namely determining whether the current traffic scenario meets the conditions for expressing driving intent based on driving environment information, includes: Acquire road topology information and motion state information of traffic participants from the driving environment information; Based on road topology information and motion state information, determine whether the current traffic scenario is a preset scenario; If the current traffic scenario is a preset scenario, then the conditions for expressing driving intent are met.
[0039] Optionally, a preset scenario refers to a specific traffic scenario type that is pre-defined and requires the vehicle to actively express its driving intentions. This scenario type is one of the prerequisites for the vehicle to trigger subsequent light signal expression processes, and its setting is based on typical interaction needs scenarios encountered by the vehicle in actual driving.
[0040] Optionally, the preset scenes include, but are not limited to, at least one of the following scene types: The scenario of an intersection without traffic lights refers to a vehicle arriving at an intersection without traffic lights, and there are vehicles or pedestrians coming from other directions. Meeting oncoming traffic in narrow road sections refers to situations where vehicles encounter oncoming or forward vehicles in narrow road sections where the lane width is insufficient for two-way traffic to pass simultaneously. A scenario where pedestrians and vehicles share a road in front of a crosswalk, where vehicles approach a crosswalk or a densely populated pedestrian area, and pedestrians are crossing the road or about to cross.
[0041] Optionally, the motion state information of traffic participants refers to the real-time kinematic data of other traffic participants in the current traffic scene that are potentially related to the vehicle's travel path, collected and identified by the vehicle perception module through cameras, lidar, millimeter-wave radar, and other sensors.
[0042] For example, the motion status information of traffic participants includes, but is not limited to, at least one of the following: Location information, namely the two-dimensional or three-dimensional spatial coordinates of the target traffic participant relative to the vehicle, can be derived from the target detection and ranging results of LiDAR point clouds or visual images. Speed information, i.e. the instantaneous speed of the target traffic participant, can be obtained from Doppler velocity measurement of millimeter-wave radar or speed estimation based on multi-frame position difference. Acceleration information, namely the magnitude and direction of the acceleration of the target traffic participant, reflects the acceleration and deceleration trend of its motion; Heading angle information, i.e. the direction of movement of the target traffic participant, is used to determine whether its driving trajectory intersects with the vehicle's driving trajectory; Motion trajectory information, i.e. the historical movement paths of a target traffic participant over a period of time, is used to predict its future driving trends and potential paths.
[0043] It should be noted that the aforementioned motion state information can be acquired in real time by the vehicle perception system and, after fusion processing, output a structured target list for the decision-making module to detect and determine path conflicts. For example, in this embodiment, when it is detected that the position of another vehicle overlaps with the expected path of this vehicle in space, and their arrival time windows intersect, it is determined that there is a traffic participant conflicting with the path of this vehicle.
[0044] In this embodiment, the light signal expression process is triggered only in specific scenarios where right-of-way is unclear, there are conflicting participants, or there is a need to express driving intentions. This reduces meaningless light signal output in scenarios where there is no need to communicate intentions, thus achieving scenario-specific triggering of light signal expression. Furthermore, since light signal expression is only triggered in preset scenarios, vehicles will not continuously generate additional light signals during normal driving, reducing visual interference or signal overload caused by excessive light output to surrounding traffic participants, and reducing unnecessary light output interference to traffic participants. Moreover, the determination of preset scenarios is automatically completed based on the road topology information and traffic participant motion status information obtained in real time by the vehicle perception system, without the need for manual judgment or operation, making the scene recognition process objective, real-time, and efficient.
[0045] In some implementations, step 102, determining whether the current traffic scenario is a preset scenario based on road topology information and motion state information, includes: Detect whether there are traffic lights or clear right-of-way signs on the current road; If there is no traffic light control and no right-of-way signs, and at least one target traffic participant that conflicts with the vehicle's path is detected, then the current traffic scenario is determined to be a preset scenario.
[0046] For example, after obtaining the aforementioned driving environment information, the vehicle's actuators, such as the intelligent driving domain controller, execute a preset scenario determination process, which may include: First, the vehicle uses the road topology information in the driving environment information to detect whether there is traffic light control at the current intersection. Specifically, the vehicle uses image data collected by the forward-facing camera to detect and identify traffic lights. The identification results show that no traffic light devices in any direction are detected. At the same time, the vehicle confirms that the intersection is a traffic light-free intersection based on high-precision map data, and further checks for the presence of clear right-of-way signs such as stop signs and yield signs. The check results also show that no such signs are found. Subsequently, based on the motion status information of traffic participants in the driving environment, the vehicle detects whether there are any target traffic participants whose travel paths conflict with its own. Specifically, the vehicle uses cameras and lidar to detect that an oncoming vehicle is approaching the intersection from the opposite lane and is located in the straight lane of the intersection. Based on the oncoming vehicle's position, speed, and heading angle, combined with the vehicle's position, speed, and expected travel path, the vehicle calculates whether the two vehicles' trajectories intersect. The calculation results show that if the vehicle enters the intersection at its current speed, the time windows for both vehicles to reach the center area of the intersection overlap, meaning their paths conflict. Based on this, the vehicle determines that there are no traffic lights or right-of-way signs at the current intersection, and that there are target traffic participants whose paths conflict with its own. Both conditions are met, so the current traffic scenario is determined to be the preset scenario (i.e., the game interaction scenario at an intersection without traffic lights), which satisfies the conditions for expressing driving intentions.
[0047] In this embodiment, road topology information and traffic participant movement status information are automatically acquired through a vehicle perception system. A dual-condition logic "AND" judgment mechanism is adopted for no traffic lights / no right-of-way signs and participants with path conflicts, which realizes the full automation of preset scene recognition. This avoids false triggering or missed triggering that may be caused by single-condition judgment, and makes the triggering of light signals focus on scenarios where there is indeed a need for traffic interaction, thereby ensuring the accuracy, rationality and scenario relevance of light signal triggering.
[0048] Step S103: In response to meeting the driving intent expression conditions, generate corresponding intent expression information based on the current traffic scenario.
[0049] Optionally, intent expression information refers to structured data generated by the vehicle decision module that characterizes the driving decision tendencies that the vehicle wishes to convey in the current traffic scenario. It is understood that intent expression information is the direct basis for subsequently determining lighting effect information, and it serves to translate the vehicle's internal decision state into semantic signals that can be recognized by the lighting control system.
[0050] Specifically, on the one hand, the intent expression information corresponds to the type of the current traffic scenario. Different preset scenarios will trigger different intent expression needs. Based on the current preset scenario type identified in step S102, such as an intersection without traffic lights, a narrow road where vehicles meet each other, or a mixed pedestrian and vehicle scenario at a pedestrian crossing, the vehicle determines the type of driving intent that may arise in that scenario from the preset mapping relationship. For example, in the scenario of an intersection without traffic lights, when the vehicle and an oncoming vehicle approach the intersection at the same time, the intent types may be "hesitation" (difficulty in determining who should go first) or "cautious passage" (deciding to go first but remaining vigilant); in the scenario of meeting each other on a narrow road, the intent types may be "actively yielding" (stopping and waiting for the other vehicle to pass) or "cautious passage" (passing slowly). That is, the scenario type determines the semantic content scope of the intent expression information. On the other hand, the intent expression information corresponds to the vehicle's real-time decision-making results in the current scenario. After determining the candidate set of intent types corresponding to the scenario, the vehicle's decision-making module evaluates various candidate driving strategies based on perception fusion data (e.g., calculating the benefit value of different decision options), and determines the intent type and its confidence parameter that most needs to be expressed at the current moment based on the evaluation results. In other words, even in the same type of preset scenario, due to differences in the motion state information of traffic participants such as position, speed, and distance, the specific content (including intent type and confidence) of the intent expression information generated by the vehicle each time may be different, thus achieving dynamic matching between the intent expression information and the current real-time traffic situation.
[0051] In some implementations, corresponding intent expression information is generated based on the current traffic scenario, including: Obtain a set of candidate driving strategies for the vehicle in the current traffic scenario; Based on the evaluation benefit value of each candidate strategy in the candidate driving strategy set, determine the type of driving intention to be expressed at present; Based on the confidence assessment results of driving intention types, generate driving intention confidence parameters corresponding to the driving intention types; The driving intention type and the corresponding driving intention confidence parameter are combined to generate intention expression information.
[0052] Optionally, the candidate driving strategy set refers to the set of all feasible driving schemes that can be selected by the vehicle decision module under the current traffic scenario. Each candidate strategy in this set corresponds to a specific driving operation scheme, including but not limited to control parameters such as acceleration magnitude, steering angle, and target speed. It is the basic set of objects that the decision algorithm must select for scheme evaluation.
[0053] For example, the generation of the candidate driving strategy set depends on the perception information of the current scene. Taking the scenario of an intersection without traffic lights as an example, the decision module can generate multiple feasible strategies in real time based on information such as the positional relationship, speed relationship, and time from the intersection between the vehicle and vehicles in the conflicting direction. The content of the candidate strategy set generated in different scenarios may be different. For example, in the scenario of meeting oncoming traffic on a narrow road, it may include strategies such as "pull over and wait", "pass slowly", and "reverse to yield", while in the scenario of a pedestrian crossing, it may include strategies such as "accelerate through", "slow down and yield to pedestrians", and "stop and wait".
[0054] Specifically, the generation methods of candidate driving strategy sets include, but are not limited to: rule-based predefined generation (i.e., pre-setting multiple candidate strategies for a specific scenario), real-time generation based on reinforcement learning policy networks (i.e., outputting multiple optional actions through decision neural networks), or parameterized sampling generation based on model predictive control (i.e., sampling multiple candidate control sequences within the feasible control space).
[0055] Optionally, the evaluation benefit value of a candidate strategy refers to the comprehensive score obtained by the decision-making module after quantitatively evaluating each candidate strategy in the candidate driving strategy set. This value is used to quantitatively measure the merits of each candidate strategy in the current traffic scenario and serves as the basic data for the decision-making module to compare strategies and determine the type of driving intent.
[0056] For example, the calculation of the evaluation benefit value employs a multi-objective weighted fusion method, including but not limited to at least one of the following: Safety benefits are based on collision risk probability assessment, such as whether the candidate strategy will cause the safe distance between the vehicle and the target traffic participant to fall below a preset threshold, and whether it may trigger emergency braking. The higher the safety score, the safer the strategy is. Traffic efficiency benefits are based on traffic time delay assessments, such as whether candidate strategies require stopping and waiting, and whether they will cause additional delays for the vehicle or other vehicles. A higher efficiency score indicates smoother traffic flow. Comfort benefits are based on vehicle motion stability assessments, such as the magnitude of jerk corresponding to candidate strategies and the absolute value of maximum acceleration. A higher comfort score indicates a better driving experience. Traffic rule compliance benefits are based on an assessment of the degree of compliance with traffic regulations, such as whether candidate strategies violate right-of-way rules.
[0057] The scores from each dimension are weighted and summed or multiplied according to different weight coefficients to obtain the comprehensive evaluation benefit value for each candidate strategy. The weight configuration of each dimension can be different in different scenarios; for example, in security-sensitive scenarios, the weight of security benefits can be increased. In this embodiment, the evaluation benefit value is normalized to a range of 0 to 1 to facilitate subsequent strategy comparison and confidence calculation.
[0058] Optionally, the confidence assessment result of the driving intention type refers to the assessment value obtained by the decision module after determining a certain driving intention type by quantitatively analyzing the degree of certainty of that intention type. The assessment result reflects the credibility of the vehicle's current judgment of the intention, specifically reflected in the distinguishability between candidate strategies.
[0059] For example, confidence assessment may be conducted in, but is not limited to, the following ways: The assessment method based on payoff difference determines the confidence level by measuring the difference between the payoff values of the optimal and suboptimal strategies. A larger difference indicates a clearer advantage of the optimal strategy, resulting in a higher confidence level for the corresponding intent type. Conversely, a smaller difference indicates less significant superiority or inferiority among the strategies, suggesting greater uncertainty in the decision-making module regarding the final selected strategy, leading to a lower confidence level. For example, the confidence level can be defined as a monotonically increasing function of the difference between the optimal and suboptimal payoff values (such as the normalized difference itself, or the difference mapped by a sigmoid function).
[0060] The evaluation method based on policy probability distribution involves converting the payoff values of each candidate policy into a probability distribution using normalization methods such as the Softmax function. The confidence level is then measured based on the information entropy or maximum probability value of this probability distribution. Higher confidence indicates a probability distribution concentrated around a single policy, while lower confidence indicates a more dispersed probability distribution.
[0061] Optionally, the driving intention confidence parameter refers to a numerical parameter generated based on the aforementioned confidence assessment results and associated with the current driving intention type. This parameter is a component of the intention expression information and serves to quantify the internal uncertainty of the decision-making module into an adjustable value, which is used for fine-tuning the dynamic parameters in the lighting effect style in subsequent steps.
[0062] In practical applications, intent expression information includes intent type and intent type confidence parameter.
[0063] Of course, in other implementations, the intent expression information may also include: scene identification information, timestamp information, etc. It is understood that scene identification information can be used for priority determination in the light language clipping logic, and timestamp information can be used for synchronized playback of light language.
[0064] In this embodiment, the intention type is determined by introducing an evaluation benefit value based on the candidate driving strategy set, and a corresponding confidence parameter is generated by combining the confidence evaluation. Finally, the intention type and confidence parameter are combined into structured intention expression information, realizing that the decision logic based on multi-strategy benefit evaluation is quantified into an identifiable intention type, so that the generation of driving intention has an objective quantitative basis. The introduction of the confidence parameter makes the intention expression information not only include the decision conclusion, but also carry the degree of decision certainty, providing a quantitative input basis for the subsequent fine-grained and gradient adjustment of lighting effects, and enriching the dimensions of intention expression. At the same time, the complex strategy evaluation results of the decision layer are summarized into a standardized data structure containing intention type and confidence parameter, realizing effective decoupling and information transmission between the decision module and the light signal control module, improving the modularity of the system architecture, and facilitating subsequent functional expansion and flexible addition and configuration of intention types.
[0065] In some implementations, the type of driving intent to be expressed is determined based on the evaluation benefit value of each candidate strategy in the candidate strategy set, including at least one of the following: When the candidate driving strategy set includes a deceleration and waiting strategy and an acceleration and passage strategy, and the difference between their evaluated benefits is less than a preset threshold, the driving intention type is determined to be a hesitant intention. When the driving behavior corresponding to the highest payoff strategy in the candidate driving strategy set is to proceed, and the difference between the highest payoff value and the payoff values of other candidate strategies is less than the second preset threshold, the driving intention type is determined to be either a cautious passage intention or a game-theoretic confirmation intention.
[0066] Optionally, the preset threshold is a preset boundary value used to determine whether the difference in benefit value between the deceleration waiting strategy and the acceleration passage strategy reaches the hesitation judgment standard. Its core function is to quantify the boundary condition where the two strategies are difficult to distinguish. That is, when the difference in benefit value between the two strategies is less than the threshold, the system determines that the vehicle is currently in a trade-off state where it is difficult to make a clear choice, thereby determining the current driving intention type as a hesitant intention.
[0067] For example, in a crossroads scenario, the payoff for a slow-down waiting strategy is 0.50, while the payoff for a speed-up strategy is 0.55, with a difference of 0.05. If the preset threshold is set to 0.10, the difference (0.05) is less than the threshold (0.10), and the system determines that the vehicle is in a hesitant state, classifying the intent as a hesitant intent. Conversely, if the difference is greater than or equal to the threshold, it indicates that the vehicle can clearly distinguish between superior and inferior strategies, and its decision-making direction is relatively clear, thus not being identified as a hesitant state.
[0068] For example, the specific value of the preset threshold can be determined through real vehicle calibration or simulation experiments, or it can be pre-configured according to factors such as driving habits in different regions and the strictness of traffic regulations.
[0069] Optionally, the second preset threshold refers to the boundary value used to determine whether the advantage of the highest-paying strategy relative to other strategies reaches the criteria for "cautious passage intention" or "game confirmation intention" when the driving behavior corresponding to the highest-paying strategy is to proceed. Its core function is to distinguish between the following two different passage decision states: Deterministic passage state: This is when the advantage of the passage strategy is very significant (the difference is greater than or equal to the second preset threshold), and the vehicle can confidently proceed. In this state, the vehicle usually does not need to express a cautious or game-theoretic intention through light signals; it can directly follow the normal passage logic or express its basic intention through conventional lights such as turn signals. Cautious / Game-Theoretic Passage State: This state occurs when the passing strategy offers the highest payoff, but its advantage is not significant (the difference is less than the second preset threshold). In other words, the passing strategy's lead over other strategies is insufficient to be considered a completely certain decision, but the vehicle is still prioritized. In this state, vehicles need to use specific light signals to convey the intention message "I'm ready to pass, but I'm still cautious / please be careful" to surrounding traffic participants. The corresponding intention type is either cautious passage intention or game-theoretic confirmation intention.
[0070] Taking the intersection scenario as an example, the reward value of the acceleration communication strategy is 0.55, the reward value of the deceleration waiting strategy is 0.5, and the reward value of other strategies is 0.52. Among them, the driving behavior corresponding to the highest reward strategy is to proceed, and the difference between the highest reward value (0.55) and the second highest reward value (0.52) is 0.03. If the second preset threshold is configured as 0.08, then the difference (0.03) is less than the threshold (0.08), and the system determines that proceeding is the priority direction, but the advantage is not significant. It is necessary to convey the intention of "proceeding cautiously" or "confirmation through game theory" to the other party through light signals to reduce the risk of the other party misinterpreting the vehicle's intention.
[0071] Understandably, the preset threshold (used to determine hesitant intent) and the second preset threshold (used to determine cautious / game-theoretic confirmation intent) can be two independent preset values, and their specific values can be the same or different. For example, they can be configured to the same value (e.g., both 0.10), or they can be configured to different values according to actual needs (e.g., the preset threshold is 0.12, and the second preset threshold is 0.08). The specific values depend on the different sensitivity requirements of the system to the difference in payoffs corresponding to the two states of "hesitation" and "caution".
[0072] In this embodiment, by setting a preset threshold and a second preset threshold respectively, the payoff difference of candidate strategies is evaluated in a graded and quantitative manner. This achieves a precise distinction between two different semantic categories: hesitant intention and cautious passage / game confirmation intention. When the payoff difference between the deceleration waiting strategy and the acceleration passage strategy is less than the preset threshold, it is determined to be a hesitant intention. When the passage strategy has the highest payoff and the difference between the highest payoff and other strategies is less than the second preset threshold, it is determined to be a cautious passage or game confirmation intention. This provides an objective quantitative basis and clear judgment boundaries for determining the type of driving intention, avoiding the ambiguity or arbitrariness of subjective settings. At the same time, the judgment logic of the two types of intentions covers a continuous decision state spectrum from "complete hesitation" to "some inclination but still needing caution", which can more delicately reflect the actual decision state of the vehicle under different game levels. This provides a more accurate and semantically clear intention input for the subsequent precise expression of light signals.
[0073] Step S104: Determine the corresponding first light effect information based on the intention expression.
[0074] Optionally, light effect information refers to a set of parameterized data generated by the light signal control module based on the intent expression information, which describes what kind of visual light effect the light-emitting device should present. This information is an intermediate data layer connecting the intent expression information and the light signal control instructions, and it is responsible for converting semantic-level driving intentions into light effect description parameters that can be recognized by the light driving hardware.
[0075] It is understandable that the light effect information is different from the final light control command. It is used to describe what kind of light effect is emitted, rather than how to drive each LEB unit at the circuit level. It can be understood that the light effect information is a parameter specification description of the light effect pattern, while the light control command is the specific circuit timing implementation that should be described.
[0076] For example, the light effect information includes, but is not limited to, at least one of the following: The basic lighting effect style identifier is used to identify the type of lighting effect mode currently being used (such as alternating sweeping lighting effect, pulse lighting effect, etc.). This identifier is derived from the basic lighting effect style found from the preset mapping relationship with the intent type as the query index. Dynamic light effect parameters are adjustable parameters used to control the dynamic performance characteristics of light effects, including but not limited to at least one of the following: sweep frequency, flicker period, brightness change rate, and pulse interval duration; Spatial distribution parameters are used to specify the spatial distribution of light effect on the light-emitting device. For example, specifying that the left and right light-emitting units are driven independently and exhibit opposite brightness changes, or specifying that the central light-emitting unit performs pulse output.
[0077] In some implementations, the dynamic light effect parameters can also be dynamically adjusted according to the intention confidence parameters generated in step S103. For example, a higher confidence value corresponds to a higher sweep frequency, and a lower confidence value corresponds to a lower sweep frequency, thereby achieving a gradient expression of the light effect.
[0078] In summary, light effect information is a complete parameterized description of "what kind of light the light-emitting device should emit," including the type identifier of the light effect, dynamic behavior parameters, and spatial distribution method, which is sufficient to support the generation of specific light signal driving control instructions in subsequent steps.
[0079] In some implementations, the first light effect information includes the sweep frequency and / or flashing period. In step S104, the corresponding first light effect information is determined based on the intention expression information, including: Extract the driving intent type and corresponding driving intent confidence parameters from the intent expression information; Using driving intent type as the query index, the system finds the basic light effect style that matches the intent type from the preset mapping relationship. The preset mapping light effect includes the correspondence between driving intent type and basic light effect style. Based on the driving intention confidence parameter, at least one dynamic parameter in the basic light effect pattern is adjusted, and the adjusted light effect pattern is used as the first light effect information; wherein, adjusting the dynamic parameter includes: controlling the sweep frequency and / or flashing period in the driving intention confidence parameter and the first light effect information to have a preset negative correlation.
[0080] For example, adjusting at least one dynamic parameter in the base lighting style based on the driving intention confidence parameter includes: Determine whether the confidence parameter of driving intention is greater than a preset first threshold; If it is greater than the first threshold, then the first frequency is used as the adjusted sweep frequency and the first period is used as the adjusted flashing period. If it is not greater than the first threshold, then the second frequency is used as the adjusted sweep frequency and the second period is used as the adjusted flashing period, wherein the second frequency is lower than the first frequency and / or the second period is longer than the first period.
[0081] In this embodiment, by extracting the driving intent type and confidence parameter from the intent expression information, the matching of basic lighting effect patterns and the adjustment of dynamic parameters are completed respectively. This achieves decoupling of intent type and confidence parameter in the lighting effect information determination process. That is, the intent type determines the semantic category of the lighting effect, and the confidence parameter determines the dynamic performance degree of the lighting effect. This allows the lighting effect information to further quantify the degree of certainty of the intent based on carrying the semantic category of the intent. At the same time, by setting the confidence parameter and the sweeping frequency and / or flashing period to a negative correlation, the dynamic visual perception of the lighting effect is associated with the confidence parameter. External traffic participants can intuitively perceive the degree of certainty of the vehicle's current decision by observing the dynamic change rhythm of the lighting effect, enriching the information carrying dimension of the light signal. Furthermore, by setting a first threshold and adopting a two-level adjustment mode, the dynamic parameter adjustment logic is simplified and standardized, reducing the system processing complexity and the computational overhead of parameter mapping, which is convenient for actual engineering deployment and real vehicle calibration.
[0082] Step S105: Based on the first light effect information, generate a light signal control command and control the vehicle's light-emitting device to execute the light signal expression based on the light signal control command.
[0083] In some implementations, step S105, generating light control instructions based on the first light effect information, includes: Obtain the first intent priority corresponding to the first light effect information; Obtain the second intent priority corresponding to the second light effect information currently being executed by the light-emitting device; Compare the priority of the first intent and the priority of the second intent; If the priority of the first intention is higher than that of the second intention, then the light control command corresponding to the first light effect information is generated; If the priority of the first intention is not higher than the priority of the second intention, then the light control command corresponding to the second light effect information is generated.
[0084] In this embodiment, an intent priority comparison mechanism is introduced to determine the intent priorities corresponding to the new and old light effect information before generating the light signal control command. When the new intent priority is higher than the currently executed intent, the new light signal is switched to be output; otherwise, the current light signal is maintained. This achieves orderly judgment of multiple intent conflicts during the execution of the light signal, ensuring that the light-emitting device outputs only one light signal at a time, avoiding signal chaos caused by multiple light effects competing for execution resources. At the same time, this mechanism ensures that high-priority intents such as safety collision avoidance can be preempted for output in a timely manner, preventing emergency information from being blocked by priority light signals, and also preventing low-priority intents from frequently interrupting the invalid switching of high-priority light signals due to slight fluctuations in perception information. This ensures the uniqueness, stability, and scene adaptability of the light signal output.
[0085] In some implementations, obtaining the intent priority includes: Obtain the first driving intention type corresponding to the first light effect information, and the second driving intention type corresponding to the second light effect information; When the first or second driving intention type is a hesitant intention, assign a priority to the pending decision type; When the first or second driving intention type is cautious passage or game-based confirmation intention, the efficiency passage class priority is assigned. The first and second intention priorities are determined according to the preset priority rules. Among them, the waiting decision category has a lower priority than the inefficient passage category.
[0086] In this embodiment, specific driving intention types (hesitation intention, cautious passage / game confirmation intention) are abstracted and categorized into a category hierarchy (decision-pending category, efficiency passage category), and a priority hierarchy relationship is established between categories (decision-pending category has lower priority than efficiency passage category). This achieves the categorization and standardization of intention priority, providing a unified categorized judgment benchmark for subsequent arbitration comparison. At the same time, this embodiment makes the decision-pending category corresponding to hesitation intentions have lower priority than the efficiency passage category corresponding to cautious passage or game confirmation intentions. This reflects a reasonable tilt in the priority ranking of efficiency passage intentions relative to pure waiting decision intentions, avoiding the frequent output of low-priority hesitation light signals due to vehicles being in a hesitant state for a long time, which would cause key passage decision signals to be submerged or delayed. This makes the light signal output more inclined to express intentions with a clear tendency to pass, which helps to guide the smooth progress of traffic flow.
[0087] In summary, the above implementation methods, by acquiring vehicle driving environment information and determining whether the current traffic scene meets the conditions for expressing driving intentions, can trigger subsequent light signal control processes when the conditions are met, achieving on-demand triggering of light signal expression rather than continuous unconditional execution. Furthermore, by generating corresponding intention expression information based on the current traffic scene and determining corresponding first light effect information based on this intention expression information, the expressed light effect is associated with the current traffic scene, enabling the vehicle to execute differentiated light signal expressions in different traffic scenarios. Further, by automatically generating light signal control commands based on the first light effect information and controlling the light-emitting device to execute the light signal expression, the entire light signal expression process is automated without manual intervention. Moreover, through the above embodiments, the vehicle can actively output light signals when specific conditions are met, realizing the proactive expression of driving intentions.
[0088] Based on the same methodological concept as the foregoing embodiments, the foregoing embodiments will be described in detail below through a specific example.
[0089] The specific embodiments of this application aim to construct a set of specific light signals to express hesitation or strategic intent in autonomous driving. The following will provide a detailed description of the system architecture, working principle, key modules, and interaction flow, and will also illustrate the core content that should be shown in each of the accompanying drawings.
[0090] I. System Overall Architecture To achieve accurate and real-time expression of interactive intentions in autonomous vehicles, this invention designs a complete system architecture. The system mainly comprises three modules, which work collaboratively to form a complete closed loop of "perception-decision-signal generation-execution," as detailed below: The Environmental Perception and Decision-Making Module includes sensors (cameras, LiDAR, millimeter-wave radar, V2X devices), high-precision maps, and decision-making algorithms (game theory models, reinforcement learning strategies). Its core function is to identify game scenarios (social scenarios without clear traffic rules) and generate interactive intentions with parameters such as confidence levels. It outputs complete intention commands containing scenario labels, intention types, and confidence levels. Scenario labels are identifiers output by the Environmental Perception and Decision-Making Module after semantically classifying the current traffic environment. They represent specific traffic situations where the vehicle needs to negotiate intentions with external participants. Examples include meeting oncoming traffic at an intersection without traffic lights, two-way traffic on a narrow road, yielding at a pedestrian crossing, merging into a roundabout, and parking space conflicts. The intention type is an interactive decision category generated by the decision-making module based on the scenario, traffic rules, and game strategies. It describes the vehicle's behavioral tendencies or negotiation stance in a specific scenario, such as right-of-way consideration (hesitation), right-of-way confirmation (I will proceed), right-of-way relinquishment (please go first), cooperative lane changing (request to insert), and route suggestion (I will drive along this curve).
[0091] Dynamic Light Signal Control Module: This is the core processing unit of the system. It has an embedded "scene-intent-light signal" mapping database and arbitration logic. It is responsible for receiving intent commands, completing light signal template queries, parameter adjustments, conflict arbitration, and finally generating specific light driving timing commands.
[0092] Dedicated vehicle exterior interactive lights: These are independent physical lights, physically isolated from traditional light groups. They are responsible for executing the light effect commands output by the dynamic light language control module and serve as the physical expression carrier of interactive intentions.
[0093] Based on this, the workflow of the light signal control module is: "input-query-parameterization-arbitration-output" to ensure the accuracy and real-time performance of the light signals. The specific steps are as follows: (1) Input: Receive the current scene label and intent command (including core parameters, such as intent confidence and scene priority) from the environment perception and decision-making module; (2) Query: Based on the scene tag and intent type, query the corresponding basic lighting effect template from the embedded "scene-intent-light language" mapping database; (3) Parameterization: Based on the parameters in the intention instruction (such as decision confidence = 0.6), adjust specific parameters of the light effect template (such as adjusting the default 1Hz sweep frequency to 0.6Hz) to achieve gradient expression of the intention; for example, gradient expression: refers to the light expression of the same intention type (such as "hesitating") is not fixed, but can change continuously and smoothly according to the continuous change of decision state parameters, so as to convey the degree of certainty of the decision more precisely and achieve a continuous visual expression from "very hesitant" to "slightly inclined".
[0094] (4) Arbitration: The built-in arbitrator checks if there is a higher-priority intent that needs to interrupt the current display. If not, proceed to step (5); if so, replace it with the light effect corresponding to the higher-priority intent to ensure signal consistency. The higher-priority intent originates from the real-time output of the environmental perception and decision-making module. When the system senses a sudden change in the environment or a decision update, it may generate a new, higher-priority intent instruction. The current display refers to the intent that the light signal control module is processing or outputting driving instructions for. The arbitrator compares the priority of the newly received intent with that of the "current display" intent to decide whether to interrupt the current light effect and immediately execute the light effect corresponding to the higher-priority intent.
[0095] (5) Output: Generate the final lighting drive timing instructions (including parameters such as on / off duration, frequency, and brightness change), send them to the dedicated vehicle exterior interactive lights, and complete the execution of the light signals.
[0096] Furthermore, the key lighting sequence design is as follows: The intersection hesitation light sequence is designed for scenarios where vehicles need to look left and right to weigh options at unsignalized intersections. It controls two independent sets of LED light strips on the left and right sides of the vehicle's front, causing their brightness to change in an inverse sine wave pattern. The period T ranges from 2 to 4 seconds (the higher the confidence level, the shorter the period; for example, a confidence level of 0.8 corresponds to a 2-second period, and a confidence level of 0.4 corresponds to a 4-second period). As the left light brightness decreases from low to high and then back to low, the right light decreases from high to low and then back to high. This dynamic effect simulates the movement of a human head looking left and right, directly conveying the hesitant intention of "weighing a choice," completely distinguishing it from the flowing light effects of existing dynamic turn signals.
[0097] The "Cautionary Passage / Confirmation" light signal sequence is used, for example, to convey the intention of "I am about to pass, please be careful" when meeting oncoming traffic on a narrow road. It controls the central LED group at the front of the vehicle to emit a unique "three flashes followed by a long pause" pattern: [300ms on → 200ms off → 300ms on → 200ms off → 300ms on → 1000ms off]. This pattern is significantly different from the uniform double flashing pattern of hazard warning lights, eliminating any possibility of confusion and ensuring extremely high recognizability.
[0098] Furthermore, conflict resolution rules can be established by setting clear intent priority rules to ensure the safety and rationality of light signals, as shown in the following examples (priority from high to low): Safety-related (emergency avoidance) > Polite interaction (proactive yielding) > Efficient passage (cautious passage) > Waiting for decision (hesitation). When the intents of "proactive yielding" and "turning" conflict, the intent of "proactive yielding" (such as the "hazard lights and nod" signal towards pedestrians) takes priority, while the turn signal remains constantly lit as the basic path indication. The two can be displayed simultaneously in space (different light groups) without conflicting or affecting information transmission.
[0099] Based on this, in a specific embodiment, taking a vehicle with intelligent driving capabilities driving to an intersection without traffic lights as an example, the technical solution of this application will be described in detail. Please refer to... Figure 2 ,like Figure 2 As shown: After the vehicle starts, the environmental perception and decision-making module initializes and then continuously integrates sensor data collected by the perception module (including cameras, LiDAR, millimeter-wave radar, etc.) to obtain real-time information about the vehicle's driving environment.
[0100] When a vehicle approaches an uncontrolled intersection, the perception module detects an oncoming vehicle approaching the intersection, and both vehicles arrive at the center of the intersection almost simultaneously. Based on road topology information (the intersection lacks traffic light control and clear right-of-way signs) and the movement status information of traffic participants (the oncoming vehicle's path conflicts with the vehicle's), the vehicle's decision-making module determines that the current traffic scenario is a preset scenario, namely a game-theoretic scenario at an uncontrolled intersection, and confirms that the conditions for expressing driving intent are met.
[0101] In response to the fulfillment of the driving intention expression conditions, the decision-making module generates corresponding intention expression information. Specifically, the decision-making module generates a strategy set containing three candidate strategies: "accelerate to pass," "decelerate to yield," and "slowly crawl and observe," and calculates the evaluation benefit values of each strategy, which are 0.55, 0.50, and 0.52, respectively. Since the difference in benefit values between the accelerate to pass strategy and the decelerate to yield strategy (0.05) is less than the preset threshold (0.10), the decision-making module determines the current driving intention type as "hesitant intention," and generates a corresponding driving intention confidence parameter of 0.45 after confidence evaluation. Finally, it combines and generates intention expression information containing the intention type and confidence parameter.
[0102] Subsequently, the light signal control module uses "hesitation intention" as the query index to retrieve the corresponding basic light effect style from the preset "scene-intention-light signal" mapping database, namely, the alternating sweeping light effect of the two sets of LED light strips on the left and right. Then, the light signal control module parametrically adjusts this basic light effect style according to the confidence parameter of 0.45: since the confidence parameter (0.45) is less than the preset first threshold (0.5), the second frequency (a lower sweeping frequency, such as 0.6Hz) and / or the second period (a longer flashing period) are used as the adjusted dynamic parameters to generate the first light effect information. Next, the light signal control module generates a light driving timing command based on the first light effect information, driving the dedicated exterior interactive lights on the lower edge of the vehicle's front bumper to execute the corresponding dynamic light signal, that is, the two sets of light strips on the left and right emit light in a slow, alternating sweeping manner, intuitively conveying to oncoming vehicles that the vehicle is weighing and observing the situation.
[0103] While the light-emitting device conveys the signal, the vehicle simultaneously performs cautious driving behavior, that is, controls the vehicle to slowly creep towards the intersection at a very low speed, so that the physical movement of the vehicle coordinates with the signal, further strengthening the intentional information conveyed to the outside world.
[0104] The vehicle continuously monitors the current scene status and changes in intent. If an oncoming vehicle slows down, stops, and flashes its lights to indicate that the vehicle should proceed first, the decision module regenerates the intent expression information based on the updated driving environment information, determining that the driving intent type is updated to "cautious passage intent" and the confidence parameter increases to 0.75. At this time, the light signal control module generates corresponding light effect information and light signal control instructions based on the new intent, controlling the light-emitting device to switch the light signal from a hesitant signal to a cautious passage signal (e.g., a pulse mode of three flashes followed by a long pause), confirming the intent "I am ready to pass" to the oncoming vehicle. If the scene ends (e.g., the vehicle has successfully passed through the intersection) or the intent no longer needs to be expressed, the light signal output stops.
[0105] Thus, the above specific embodiments clearly demonstrate the complete closed-loop control process of "perception - scene determination - intent generation - light signal query - parameter adjustment - command generation - execution - synchronized driving behavior - stop / switch".
[0106] In the above embodiments of this application, the vehicle perception system acquires driving environment information in real time and automatically identifies preset game scenarios, realizing on-demand triggering of light signals in scenarios where driving intentions need to be communicated externally; by mapping the intention expression information containing intention type and confidence parameters generated by the decision module to the basic light effect style and generating corresponding light effect information and light signal control instructions after parameterization, the association and correspondence between light signals and the current traffic scenario is realized, so that different driving intention types correspond to different light effect styles, and the dynamic performance of the same intention type also differs due to different confidence levels, thereby enabling the outside world to perceive the vehicle's decision-making state through light effect styles and dynamic rhythms; by automatically generating driving timing instructions through the light signal control module and controlling the dedicated light-emitting device to execute light effects, the fully automated control of light signals is realized without manual intervention; by synchronously executing cautious driving behavior that matches the intention, the coordination between light signals and vehicle physical movement behavior is realized, further enhancing the consistency of external intention communication; by continuously monitoring the scenario status and stopping or switching light signals when the scenario ends or the intention changes, the adaptive adjustment of light signals with the dynamic changes of the traffic scenario is realized.
[0107] Based on the same inventive concept as the foregoing embodiments, this invention provides a vehicle, such as... Figure 3 As shown, the vehicle includes: a processor 410 and a memory 411 storing a computer program; wherein, Figure 3 The processor 410 shown in the diagram does not indicate that there is only one processor 410, but only indicates the positional relationship of the processor 410 relative to other devices. In practical applications, there can be one or more processors 410; similarly, Figure 3 The memory 411 shown in the diagram has the same meaning, that is, it is only used to indicate the positional relationship of memory 411 relative to other devices. In practical applications, there can be one or more memories 411. When the processor 410 runs the computer program, the above-mentioned vehicle light signal control method is implemented.
[0108] The vehicle may also include at least one network interface 412. Various components in the vehicle are coupled together via a bus system 413. It is understood that the bus system 413 is used to enable communication between these components. In addition to a data bus, the bus system 413 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 413.
[0109] The memory 411 can be volatile or non-volatile, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 411 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.
[0110] The memory 411 in this embodiment of the invention is used to store various types of data to support the operation of the vehicle. Examples of this data include: any computer programs used to operate on the vehicle, such as operating systems and applications; contact data; phonebook data; messages; pictures; videos, etc. The operating system includes various system programs, such as a framework layer, core library layer, driver layer, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications, such as media players, browsers, etc., used to implement various application services. Here, the program implementing the method of this embodiment of the invention can be included in the application.
[0111] Based on the same inventive concept as the foregoing embodiments, this embodiment also provides a machine-readable storage medium storing a computer program. The machine-readable storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it can also be various devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc. When the computer program stored in the machine-readable storage medium is executed by a processor, it implements a vehicle light signal control method applied to the aforementioned vehicle. For the specific steps implemented when the computer program is executed by the processor, please refer to [link to relevant documentation]. Figure 1 , Figure 2 The description of the illustrated embodiments will not be repeated here.
[0112] 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.
[0113] In this document, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, which includes not only the elements listed but also other elements not expressly listed.
[0114] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for controlling vehicle lighting signals, characterized in that, include: Obtain vehicle driving environment information; Based on the driving environment information, determine whether the current traffic scenario meets the conditions for expressing driving intent; In response to satisfying the driving intention expression conditions, corresponding intention expression information is generated based on the current traffic scenario; Based on the expressed intent information, determine the corresponding first light effect information; Based on the first light effect information, a light signal control command is generated, and the vehicle's light-emitting device is controlled to execute the light signal expression based on the light signal control command.
2. The method according to claim 1, characterized in that, The step of determining whether the current traffic scenario meets the conditions for expressing driving intent based on the driving environment information includes: Obtain the road topology information and the motion state information of traffic participants from the driving environment information; Based on the road topology information and the motion state information, determine whether the current traffic scenario is a preset scenario; If the current traffic scenario is the preset scenario, then the driving intention expression condition is satisfied.
3. The method according to claim 2, characterized in that, The step of determining whether the current traffic scenario is a preset scenario based on the road topology information and the motion state information includes: Detect whether there are traffic lights or clear right-of-way signs on the current road; If there is no traffic light control and no right-of-way sign, and at least one target traffic participant that conflicts with the path of the vehicle is detected, then the current traffic scenario is determined to be the preset scenario.
4. The method according to claim 1, characterized in that, The step of generating corresponding intent expression information based on the current traffic scenario includes: Obtain the set of candidate driving strategies for the vehicle in the current traffic scenario; Based on the evaluation benefit value of each candidate strategy in the candidate driving strategy set, determine the type of driving intention to be expressed. Based on the confidence assessment results of the driving intention type, a driving intention confidence parameter corresponding to the driving intention type is generated; The driving intention type and the corresponding driving intention confidence parameter are combined to generate the intention expression information.
5. The method according to claim 4, characterized in that, The step of determining the type of driving intention to be expressed based on the evaluation benefit value of each candidate strategy in the candidate driving strategy set includes at least one of the following: When the candidate driving strategy set includes a deceleration and waiting strategy and an acceleration and passage strategy, and the difference between their evaluated benefits is less than a preset threshold, the driving intention type is determined to be a hesitant intention. When the driving behavior corresponding to the highest-paying strategy in the candidate driving strategy set is to proceed, and the difference between the highest-paying value and the paying values of other candidate strategies is less than a second preset threshold, the driving intention type is determined to be either a cautious passage intention or a game-theoretic confirmation intention.
6. The method according to claim 4, characterized in that, The first light effect information includes the sweep frequency and / or flashing period; determining the corresponding first light effect information based on the intention expression information includes: Extract the driving intention type and the corresponding driving intention confidence parameter from the intention expression information; Using the driving intention type as a query index, a basic light effect style that matches the intention type is found from a preset mapping relationship, wherein the preset mapping relationship includes the correspondence between driving intention types and the basic light effect styles; Based on the driving intention confidence parameter, at least one dynamic parameter in the basic light effect pattern is adjusted, and the adjusted light effect pattern is used as the first light effect information; wherein, adjusting the dynamic parameter includes: controlling the driving intention confidence parameter to have a preset negative correlation with the sweeping frequency and / or flashing period in the first light effect information.
7. The method according to claim 1, characterized in that, The step of generating light control commands based on the first light effect information includes: Obtain the first intent priority corresponding to the first light effect information; Obtain the second intent priority corresponding to the second light effect information currently being executed by the light-emitting device; Compare the priority of the first intent with the priority of the second intent; If the priority of the first intention is higher than the priority of the second intention, then a light control command corresponding to the first light effect information is generated; If the priority of the first intention is not higher than the priority of the second intention, then a light control instruction corresponding to the second light effect information is generated.
8. The method according to claim 7, characterized in that, The acquisition of the intent priority includes: Obtain the first driving intention type corresponding to the first light effect information, and the second driving intention type corresponding to the second light effect information; When the first driving intention type or the second driving intention type is a hesitant intention, a waiting decision-making priority is assigned; When the first driving intention type or the second driving intention type is a cautious passage intention or a game-theoretic confirmation intention, the efficiency passage class priority is assigned. The first intent priority and the second intent priority are determined according to a preset priority rule, wherein the priority of the waiting decision class is lower than the priority of the efficiency passage class.
9. A vehicle, characterized in that, include: A processor and a memory for storing executable instructions; wherein the processor is configured to execute the instructions to implement the vehicle light signal control method as described in any one of claims 1-8.
10. A machine-readable storage medium, characterized in that, When the instructions in the machine-readable storage medium are executed by a processor, the vehicle light signal control method as described in any one of claims 1-8 is implemented.
