Driving mode determination method and device, vehicle, storage medium and computer program

Abstract

The present disclosure relates to a driving mode determination method, device, vehicle, storage medium and computer program. It relates to the technical field of intelligent driving. The method comprises: in response to a network state being an abnormal state, determining an offline driving distance of a vehicle; determining a network recovery probability by acquiring network state data; and determining a driving mode of the vehicle according to the offline driving distance and the network recovery probability. The driving mode determination method of the present disclosure determines the driving mode corresponding to the vehicle by combining the offline driving distance of the vehicle and the network recovery probability after the network is in an abnormal state. The driving mode is determined by multiple conditions, so that the driving mode most suitable for the current network state can be determined when the network is abnormal, the stability of the driving mode switching is improved, the continuity of the route when the network is abnormal is solved, and the road traffic safety is improved.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent driving technology, and more specifically, to a driving mode determination method, a driving mode determination device, a vehicle, a computer-readable storage medium, and a computer program. Background Technology

[0002] With the rapid development of new energy vehicles, driver assistance technology has increasingly become a core focus of the industry, especially advanced navigation-assisted driving functions which rely heavily on online networks. Currently, mainstream driver assistance systems generally employ online maps and real-time data processing to achieve functions such as lane-level navigation, road condition updates, new energy vehicle power planning, and closed-loop control for driver assistance. However, in actual driving, vehicles often experience network interruptions when entering tunnels, mountainous areas, or areas with weak signal coverage, rendering online navigation services unavailable. In such cases, the inability to obtain online data in a timely manner can lead to inaccurate routes, downgraded or even discontinued driver assistance functions, seriously impacting driving safety and user experience. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this disclosure provides a driving mode determination method, a driving mode determination device, a vehicle, a computer-readable storage medium, and a computer program.

[0004] According to a first aspect of the present disclosure, a driving mode determination method is provided, the method comprising: determining the offline driving distance of a vehicle in response to an abnormal network state; determining a network recovery probability by acquiring network state data; and determining the driving mode of the vehicle based on the offline driving distance and the network recovery probability.

[0005] In some exemplary embodiments of this disclosure, determining the offline driving distance of a vehicle in response to an abnormal network state includes: in response to the abnormal network state, acquiring the driving time and driving speed of the vehicle; and determining the offline driving distance based on the driving time and driving speed.

[0006] In some exemplary embodiments of this disclosure, the step of determining the network recovery probability by acquiring network status data includes: acquiring the network status data; acquiring historical network status data; and determining the network recovery probability based on the network status data and the historical network status data.

[0007] In some exemplary embodiments of this disclosure, obtaining the network status data includes: in response to the network status being in a connected state, obtaining the network status data based on a first monitoring frequency; and in response to the network status being in the abnormal state, obtaining the network status data based on a second monitoring frequency, wherein the second monitoring frequency is higher than the first monitoring frequency.

[0008] In some exemplary embodiments of this disclosure, determining the driving mode of the vehicle based on the offline driving distance and the network recovery probability includes: inputting the offline driving distance and the network recovery probability into a pre-trained driving mode classification model to obtain the driving mode; the driving mode classification model is trained through the following steps: acquiring model training data, the model training data including model input features and driving mode classification results, the model input features including network recovery probability data and offline driving distance data; training an initial model using the model training data to obtain the driving mode classification model.

[0009] In some exemplary embodiments of this disclosure, determining the driving mode of the vehicle based on the offline driving distance and the network recovery probability includes: determining the driving mode as a first mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being greater than or equal to a recovery probability threshold; acquiring offline cached data in response to the driving mode being the first mode; generating a temporary navigation planning scheme for the vehicle based on the offline cached data; acquiring first online data in response to the network status being connected; and generating a driving plan for the vehicle based on the first online data and the offline cached data.

[0010] In some exemplary embodiments of this disclosure, determining the driving mode of the vehicle based on the offline driving distance and the network recovery probability further includes: determining the driving mode as a second mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being less than a recovery probability threshold; acquiring offline cache data in response to the driving mode being the second mode; updating the offline driving path of the vehicle using the offline cache data; and generating a driving plan for the vehicle based on the updated offline driving path.

[0011] In some exemplary embodiments of this disclosure, determining the driving mode of the vehicle based on the offline driving distance and the network recovery probability further includes: determining the driving mode as a third mode in response to the offline driving distance being greater than an offline driving distance threshold; sending a warning message to the user in response to the driving mode being the third mode, the warning message including that the vehicle has exited the assisted driving mode; acquiring offline cached data and determining the vehicle's driving path information based on the offline cached data; and returning the start time of the third mode and the switching information of the driving mode.

[0012] In some exemplary embodiments of this disclosure, obtaining offline cached data includes: obtaining second online data, wherein the second online data is online data obtained before the network state is detected to be in an abnormal state; determining a timestamp marker corresponding to the second online data; in response to the network state being in the abnormal state, determining a valid time value corresponding to the second online data; and determining the offline cached data from the second online data based on the vehicle location, the valid time value, and the timestamp marker.

[0013] In some exemplary embodiments of this disclosure, the second online data includes one or more of the following: lane-level navigation data, traffic snapshot data, estimated arrival time, battery information, and assisted driving closed-loop parameters.

[0014] In some exemplary embodiments of this disclosure, determining the offline cached data from the second online data based on the vehicle location, the valid time value, and the timestamp includes: dynamically cropping the second online data based on the vehicle location, the valid time value, and the timestamp to obtain the offline cached data.

[0015] According to a second aspect of the present disclosure, a driving mode determination device is provided, comprising: an offline distance determination unit, configured to determine the offline driving distance of a vehicle in response to an abnormal network state; a network recovery prediction unit, configured to determine a network recovery probability by acquiring network state data; and a driving mode determination unit, configured to determine the driving mode of the vehicle based on the offline driving distance and the network recovery probability.

[0016] In some exemplary embodiments of this disclosure, the driving mode determination unit includes a driving mode determination subunit; the driving mode determination subunit is used to input the offline driving distance and the network recovery probability into a pre-trained driving mode classification model to obtain the driving mode; the driving mode classification model is trained through the following steps: acquiring model training data, the model training data including model input features and driving mode classification results, the model input features including network recovery probability data and offline driving distance data; training an initial model using the model training data to obtain the driving mode classification model.

[0017] In some exemplary embodiments of this disclosure, the driving mode determination unit includes at least a first mode determination unit, a cached data acquisition unit, a temporary navigation generation unit, an online data acquisition unit, and a first scheme generation unit; the first mode determination unit is configured to determine the driving mode as a first mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being greater than or equal to a recovery probability threshold; the cached data acquisition unit is configured to acquire offline cached data in response to the driving mode being the first mode; the temporary navigation generation unit is configured to generate a temporary navigation planning scheme for the vehicle based on the offline cached data; the online data acquisition unit is configured to acquire first online data in response to the network status being in a connected state; and the first scheme generation unit is configured to generate a driving scheme for the vehicle based on the first online data and the offline cached data.

[0018] In some exemplary embodiments of this disclosure, the driving mode determination unit includes at least a second mode determination unit, a cached data acquisition unit, and an offline path optimization unit; the second mode determination unit is used to determine the driving mode as a second mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being less than a recovery probability threshold; the cached data acquisition unit is used to acquire offline cached data in response to the driving mode being the second mode; the offline path optimization unit is used to update the offline driving path of the vehicle using the offline cached data, and generate a driving plan for the vehicle based on the updated offline driving path.

[0019] In some exemplary embodiments of this disclosure, the driving mode determination unit includes at least a third mode determination unit, a warning notification unit, a cached data acquisition unit, and an information feedback unit; the third mode determination unit is used to determine the driving mode as a third mode in response to the offline driving distance being greater than an offline driving distance threshold; the warning notification unit is used to send a warning notification to the user in response to the driving mode being the third mode, the warning notification including that the vehicle has exited the assisted driving mode; the cached data acquisition unit is used to acquire offline cached data and determine the vehicle's driving path information based on the offline cached data; the information feedback unit is used to return the start time of the third mode and the switching information of the driving mode.

[0020] In some exemplary embodiments of this disclosure, the cached data acquisition unit includes a cached data acquisition subunit; the cached data acquisition subunit is configured to acquire second online data, the second online data being online data acquired before the network state is detected to be in an abnormal state; determine a timestamp marker corresponding to the second online data; in response to the network state being in the abnormal state, determine a valid time value corresponding to the second online data; and determine the offline cached data from the second online data based on the vehicle location, the valid time value, and the timestamp marker.

[0021] In some exemplary embodiments of this disclosure, the cached data acquisition unit includes a cached data acquisition subunit, which is used to dynamically trim the second online data according to the vehicle location, the valid time value and the timestamp mark to obtain the offline cached data.

[0022] According to a third aspect of the present disclosure, a vehicle is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to: implement the steps of the driving mode determination method according to any one of the present disclosures.

[0023] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of a mobile terminal, the mobile terminal is enabled to perform any of the driving mode determination methods described herein.

[0024] According to a fifth aspect of the present disclosure, a computer program is provided, characterized in that it includes a computer program that, when executed by a processor, implements any of the driving mode determination methods described in the present disclosure.

[0025] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: The driving mode determination method disclosed herein, when the network is in an abnormal state, determines the corresponding driving mode for the vehicle by combining the vehicle's offline driving distance and the probability of network recovery. By determining the driving mode through multiple conditions, the most suitable driving mode for the current network state can be selected when the network is abnormal, improving the stability of driving mode switching, solving the problem of route continuity when the network is abnormal, and enhancing road traffic safety.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0028] Figure 1 This is a flowchart illustrating a driving mode determination method according to an exemplary embodiment of the present disclosure.

[0029] Figure 2 This is a flowchart illustrating a process for determining a vehicle's status based on its offline driving distance, according to an exemplary embodiment of this disclosure.

[0030] Figure 3 This is a flowchart illustrating the acquisition of network status data and determination of network recovery probability according to an exemplary embodiment of the present disclosure.

[0031] Figure 4 This is a flowchart illustrating the determination of a vehicle's driving mode according to an exemplary embodiment of this disclosure.

[0032] Figure 5 This is a flowchart illustrating the acquisition of offline cached data according to an exemplary embodiment of the present disclosure.

[0033] Figure 6 This is a flowchart illustrating the generation of driving schemes based on different driving modes according to an exemplary embodiment of the present disclosure.

[0034] Figure 7 This is a system architecture diagram illustrating an implementation of a driving mode determination method according to an exemplary embodiment of the present disclosure.

[0035] Figure 8 This is a timing diagram illustrating a driving mode determination method according to an exemplary embodiment of the present disclosure.

[0036] Figure 9 This is a block diagram illustrating a driving mode determination device according to an exemplary embodiment of the present disclosure.

[0037] Figure 10 This is a functional block diagram of a vehicle according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0038] Exemplary embodiments of this disclosure will be described in detail herein, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a particular order. Furthermore, for clarity and brevity, descriptions of features known in the art may be omitted.

[0039] The embodiments described below, which are examples of some of the embodiments of this disclosure, do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0040] With the development of car navigation technology and the widespread use of unlimited data cards for in-vehicle systems, the application scenarios for offline navigation are gradually narrowing. However, in the fields of new energy navigation, lane-level navigation, and assisted driving, online capabilities have become the core reliance, and offline data has significant shortcomings, such as not supporting lane-level navigation, traffic updates, multiple route alternatives, and closed-loop parameter updates for assisted driving systems such as Navigate on Autopilot (NOA) or Highway Navigate and Autopilot (HNOA) for highways.

[0041] In assisted driving scenarios, after a network outage, offline routes cannot dynamically adapt to real-time traffic conditions, lane-level information, and new energy vehicle power planning, resulting in insufficient route accuracy. Because related solutions lack monitoring and response mechanisms for network recovery timeliness, they cannot prioritize route recovery within a specified critical distance, easily causing the system to degrade to Lane Centering Control (LCC) mode. Data gaps exist during the switching between offline and online routes, such as the loss of HNOA (Hardware-Oriented Autopilot) closed-loop parameters, leading to unstable driving modes after the switch. The loss of advanced assisted driving capabilities severely impacts driving safety, route accuracy, and user experience. However, existing technologies cannot solve the route recovery problem after network anomalies in assisted driving scenarios. Therefore, a method for route recovery within a critical distance of offline vehicle travel is needed.

[0042] The steps of the method in the exemplary embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings and examples.

[0043] Figure 1 This is a flowchart illustrating a driving mode determination method according to an exemplary embodiment of the present disclosure. The method of this embodiment can be applied to electronic devices, including vehicles.

[0044] like Figure 1 As shown, in some embodiments, the driving mode determination method of this disclosure includes: In step S110, in response to an abnormal network status, the offline driving distance of the vehicle is determined.

[0045] In this embodiment of the disclosure, various data acquisition devices and network testing tools can be installed on the vehicle. These devices and tools can be used to sense and acquire the network status used by the vehicle during navigation, allowing for a determination of whether to switch the vehicle's driving mode based on the network status.

[0046] In this embodiment, network status monitoring is performed using a network testing tool installed in the vehicle. The network connection status is monitored in real time, and the network connected to the vehicle can include, but is not limited to, 4G, 5G, and WiFi. When an abnormal network status is detected, the cumulative distance traveled by the vehicle during this abnormal network state is determined, i.e., the offline travel distance. The offline travel distance can be the cumulative distance traveled after the abnormal network status was detected. An abnormal state can be a situation where the network cannot obtain planning data for vehicle navigation; abnormal states can include, but are not limited to, network interruption, weak network signal preventing the acquisition of navigation planning data, and unstable network preventing the acquisition of navigation planning data.

[0047] In step S120, the network recovery probability is determined by acquiring network status data.

[0048] In vehicle driving scenarios, besides the vehicle's offline driving distance, network recovery probability is also a crucial factor in determining the vehicle's driving mode. Therefore, when the network is in an abnormal state, the network recovery probability can be determined by acquiring network status data. Network status data can include network parameters such as Reference Signal Receiving Power (RSRP), Signal-to-Interference Plus Noise Ratio (SINR), data transmission rate, latency, and packet loss rate, acquired by data acquisition equipment after the network is in an abnormal state.

[0049] After acquiring network status data, the network recovery probability is predicted through time series analysis. The network recovery probability can be the probability that the network will recover to a normal connection state within a specified time or a specified critical distance in the future. The normal connection state can be the state in which the network can continuously acquire the online data (such as vehicle navigation planning data) required by the vehicle in assisted driving mode during vehicle navigation.

[0050] In step S130, the driving mode of the vehicle is determined based on the offline driving distance and the network recovery probability.

[0051] The vehicle's driving mode can be determined based on the offline driving distance and the network recovery probability. For example, when the network is connected, the vehicle uses an assisted driving mode, such as NOA mode or HNOA mode. When the vehicle's offline driving distance exceeds a critical distance threshold or the network recovery probability falls below a preset recovery probability threshold, NOA or HNOA mode can no longer be maintained. At this point, the vehicle's driving mode needs to be switched, downgrading the current high-level NOA or HNOA mode to a low-level driving mode, exiting the assisted driving mode, and subsequent navigation planning will be based on the switched driving mode.

[0052] If the vehicle's offline driving distance and network recovery probability meet the requirements for maintaining the current driving mode, there is no need to downgrade the mode. The current NOA or HNOA mode will still be used for navigation planning, avoiding a direct downgrade to a lower-level driving mode after the network is in an abnormal state, thus improving the continuity and safety of assisted driving.

[0053] The driving mode determination method disclosed herein, when the network is in an abnormal state, determines the corresponding driving mode for the vehicle by combining the vehicle's offline driving distance and the probability of network recovery. By determining the driving mode through multiple conditions, the most suitable driving mode for the current network state can be selected when the network is abnormal, improving the stability of driving mode switching, solving the problem of route continuity when the network is abnormal, and enhancing road traffic safety.

[0054] Figure 2 This is a flowchart illustrating a process for determining a vehicle's state based on its offline driving distance, according to an exemplary embodiment of this disclosure. Figure 1 Based on the driving mode determination method shown, Figure 1 Step S110 shown may include the following steps.

[0055] In step S210, a network anomaly is triggered. A network anomaly is triggered when the network testing tool detects an abnormal network state. In step S220, the time of the network anomaly is recorded. For example, the time when the network anomaly is detected is designated as the first time point, denoted as t1. In step S230, the vehicle's position, speed, and travel time are acquired. The data acquisition device in the vehicle can acquire information such as the vehicle's position, speed v, and the second time point t2. Based on the second time point t2 and the first time point t1, the vehicle's travel time t can be determined.

[0056] In step S240, the offline driving distance is calculated based on the driving speed and driving time, i.e., offline driving distance = driving speed v × driving time t. Since the vehicle's driving speed may change dynamically, the driving time at different driving speeds can be calculated separately, and the resulting driving distances can be summed to obtain the offline driving distance. In another embodiment, the offline driving distance can also be determined based on the first vehicle position when the network is detected to be in an abnormal state, and the second vehicle position obtained after the network abnormality.

[0057] In step S250, the offline driving distance is compared with an offline driving distance threshold. In step S260, if the offline driving distance is less than or equal to the offline driving distance threshold, the vehicle state is determined to be in the first state (critical warning state). Otherwise, in step S270, if the offline driving distance is greater than the offline driving distance threshold, the vehicle state is determined to be in the second state (degradation trigger state); and in step S280, LCC mode degradation preparation is triggered. By obtaining the vehicle's driving speed and driving time, the vehicle's offline driving distance can be determined, and the offline driving distance is an important decision factor in determining the vehicle's driving mode.

[0058] Figure 3 This is a flowchart illustrating the acquisition of network status data and determination of network recovery probability according to an exemplary embodiment of this disclosure. Figure 1 Based on the driving mode determination method shown, Figure 1 Step S120 shown may include the following steps.

[0059] In step S310, network monitoring is initiated. In a vehicle driving scenario, network status can be continuously monitored using network monitoring tools within the vehicle; for example, network connection status can be determined by acquiring network status data. Acquiring network status data can be achieved through the following steps.

[0060] In step S320, it is determined whether the network is in a normal connection state. If the network is in a normal connection state, then in step S330, network status data is acquired at a first monitoring frequency. For example, the first monitoring frequency could be once every 10 seconds. If the network is in an abnormal state, then in step S340, network status data is acquired at a second monitoring frequency. For example, the second monitoring frequency could be once per second.

[0061] In step S350, network status data is collected. The data acquisition tool can obtain network status data at different monitoring frequencies under different network conditions. For example, when the network is in an abnormal state, the network status parameters collected at a second monitoring frequency are used as network status data and are then used in the subsequent prediction of network recovery probability.

[0062] It will be readily understood by those skilled in the art that, in some other embodiments of this disclosure, the specific values ​​of the first monitoring frequency and the second monitoring frequency can be set according to network monitoring requirements. This disclosure does not impose any special limitations on this, but usually the second monitoring frequency needs to be higher than the first monitoring frequency, so that when the network is in a normal state, a lower frequency is used for collection to avoid wasting resources; at the same time, when the network is in an abnormal state, the network recovery dynamics can be quickly captured.

[0063] By acquiring network status data, not only can the network status be determined, but the probability of network recovery can also be determined. In step S360, the probability of network recovery is predicted. The network recovery probability can provide a basis for decision-making regarding offline / online route switching and adjustment of assisted driving modes.

[0064] The probability of network recovery can be determined based on network status data and historical network status data. After obtaining the network status data, historical network status data is acquired. This historical network status data can be historical data related to the network status data collected through network monitoring tools. In addition to data related to network status parameters, historical network status data also includes two types of correlated data: one is historical network data under similar scenarios, such as data related to signal fluctuation patterns in the same road segment, time period, and weather; the other is the full-cycle data of the historical network from interruption to recovery, such as the parameter threshold at the time of historical network interruption, the average time from that threshold to network recovery, and the corresponding recovery probability.

[0065] Time series analysis is used to process acquired network state data and historical network state data to predict network recovery probabilities, such as determining the probability of network recovery within a specified future timeframe (e.g., within 10 seconds or when a vehicle travels a specified distance). In calculating the network recovery probability, a mathematical model is established, with the collected parameters used as variables. Machine learning algorithms or statistical methods are then employed to analyze the relationship between parameter changes and network recovery, thereby obtaining the network recovery probability. This network state data is used collectively to improve the accuracy of network recovery probability predictions, avoiding misjudgments that can result from relying solely on network state data.

[0066] In step S370, the network status and network recovery probability are output. The network recovery probability can be represented by a specific numerical value, such as 30%, 60%, or 75%; it can also be represented by different levels, such as high, medium, and low, or other numbers of levels. For example, a specific percentage can be calculated through time series analysis, such as 60% corresponding to "medium" and 80% corresponding to "high." This retains precise numerical values ​​for the system's backend decision-making (such as triggering pre-switch preparation when the probability is 60%), and can also map the numerical values ​​to corresponding levels for simplified presentation or rapid judgment.

[0067] exist Figure 1 Based on the driving mode determination method shown, Figure 1 Step S130 shown can be implemented through the following steps: Obtain a pre-trained driving mode classification model, input the offline driving distance and network recovery probability into the driving mode classification model, and output the corresponding driving mode through the driving mode classification model.

[0068] The driving mode classification model can be trained through the following steps: Acquiring model training data. In this embodiment, network recovery probability data and offline driving distance data are used as model input features. The network recovery probability data can be multiple different recovery probabilities representing whether the network can recover from an abnormal state to a normal connection state. For example, the network recovery probability data can include the critical probability corresponding to the network recovery from an abnormal state. The offline driving distance data can be multiple different offline driving distances representing whether the network can recover from an abnormal state to a normal connection state. For example, the offline driving distance data can include the critical offline driving distance corresponding to the network recovery from an abnormal state to a normal connection state.

[0069] The driving mode classification result corresponding to the model input features is determined. This result can be the driving mode corresponding to different values ​​of network recovery probability and offline driving distance. The model input features and their corresponding driving mode classification results are combined to form a training dataset, thus constructing a training dataset containing multiple training datasets. The initial model is trained using these multiple training datasets to obtain the driving mode classification model. The initial model can be implemented based on various classification model structures. For example, it can be implemented using rule-based decision algorithms or machine learning algorithms.

[0070] For the decision tree algorithm, network recovery probability and distance critical state are used as features. Distance critical state can be the vehicle state corresponding to different offline driving distances. The first mode (such as triggering route recovery), the second mode (such as initiating offline route optimization), and the third mode (such as triggering LCC downgrade) are used as different driving mode classification results. The decision tree is trained with a large amount of historical data to learn the best decision to be made under different feature combinations, thus obtaining the driving mode classification model.

[0071] Alternatively, reinforcement learning algorithms can be used to treat the processing module that determines the driving mode as an agent. This agent selects different actions (such as triggering route recovery) based on the current network recovery probability and distance threshold. It receives rewards or penalties based on the results of these actions (such as whether navigation is successfully maintained). Through continuous interaction with the environment, it learns the optimal decision-making strategy, and the agent after reinforcement learning serves as the driving mode classification model. By linking network monitoring with distance monitoring, the driving mode can be determined within a specified offline driving distance, prioritizing route recovery rather than directly downgrading to a lower-level driving mode, thus reducing the probability of system degradation.

[0072] In some other embodiments of this disclosure, other network models can also be used to construct the driving mode classification model, such as logistic regression, support vector machine (SVM), random forest and neural network. This disclosure does not impose any special limitations on the model structure used in the driving mode classification model.

[0073] In addition, this disclosure also provides a rule-based decision-making algorithm for determining the driving mode of a vehicle. Figure 4 This is a flowchart illustrating the determination of a vehicle's driving mode according to an exemplary embodiment of this disclosure. Figure 1 Based on the driving mode determination method shown, Figure 1 As shown in step S130, the first mode can be determined through the following steps.

[0074] In step S410, the network recovery probability and offline driving distance are received. In step S420, it is determined whether the offline driving distance is less than or equal to an offline driving distance threshold. The offline driving distance threshold can be the longest distance that NOA mode can maintain assisted driving without degrading without inputting online data; for example, the offline driving distance threshold can be configured to 3 kilometers.

[0075] When determining the offline driving distance threshold, considering the accuracy requirements and practical applications of the driver assistance system, appropriate error compensation measures are typically implemented. Distance monitoring is based on calculations using the Global Positioning System (GPS) and vehicle speed. However, GPS itself has inherent errors, and vehicle speed sensors may also have accuracy deviations. To reduce these errors, multi-sensor fusion technology can be used, combining data from other sensors such as the Inertial Navigation System (INS) to correct the distance calculation results. Alternatively, map matching algorithms can be employed to match the vehicle's trajectory with road information on electronic maps, optimizing the distance calculation results based on road shape and mileage information to determine the offline driving distance threshold.

[0076] If the offline driving distance is less than or equal to the offline driving distance threshold, then step S430 is executed to determine whether the network recovery probability is less than or equal to the recovery probability threshold. The recovery probability threshold can be determined by statistically analyzing the proportion of online route recovery triggered under different network recovery probabilities and successfully maintaining normal navigation. For example, the recovery probability threshold can be configured to 60%.

[0077] The recovery probability threshold is calculated through the following steps: The recovery probability threshold can be obtained through actual testing and simulation experiments. By statistically analyzing the proportion of online route recovery triggered and normal navigation successfully maintained under different network recovery probabilities, and comprehensively considering factors such as system performance, security, and stability, it is determined that when the recovery probability reaches the recovery probability threshold, the feasibility of network recovery can be guaranteed to a large extent, and the risk of system degradation can be effectively reduced within the critical distance of the offline driving distance threshold. Therefore, it is set as the threshold for triggering online route recovery.

[0078] It will be readily understood by those skilled in the art that both the offline driving distance threshold and the recovery probability threshold can be configured according to the actual driving needs of the vehicle, thereby achieving the purpose of deciding the driving mode, reducing the probability of driving mode downgrading, and maintaining the continuity of advanced driver assistance. This disclosure does not impose any special limitations on the specific values ​​of the offline driving distance threshold and the recovery probability threshold.

[0079] If the network recovery probability is greater than or equal to the recovery probability threshold, then in step S440, the driving mode is determined as the first mode, also known as the online route recovery mode, and online route recovery is triggered. The first mode can be a driving mode that triggers online route recovery. In the first mode, online route recovery will be triggered first, and a driving plan for the vehicle will be generated using the acquired offline cached data.

[0080] After setting the driving mode to the first mode, the vehicle's driving plan generated from offline cached data can be achieved through the following steps: The offline cached data is used as "temporary transitional data." Before the network recovers and online data is loaded, a temporary navigation plan for the vehicle is generated based on the offline cached data. This temporary navigation plan is used to perform temporary navigation planning for the vehicle, maintaining the continuity of assisted driving navigation until the network recovers and online data is loaded.

[0081] When the network returns to normal, the data acquisition device will continuously acquire the first online data through the network. After the first online data is loaded, it will integrate the offline cached data to generate the vehicle's driving plan, achieving seamless switching and avoiding route gaps.

[0082] Figure 5 This is a flowchart illustrating the acquisition of offline cached data according to an exemplary embodiment of this disclosure. Figure 4Based on the driving mode determination method shown, Figure 4 As shown in step S440, the offline cache data used to trigger online route recovery can be obtained through the following steps.

[0083] In step S510, if the network is in a normal state, step S520 is executed to continuously cache second online data. This second online data may include, but is not limited to, lane-level navigation data, traffic snapshot data, estimated arrival time, battery information, and assisted driving closed-loop parameters. The second online data can be the core data required to maintain the vehicle in assisted driving mode. Continuously caching this second online data allows the vehicle to maintain assisted driving mode.

[0084] Lane-level navigation data may include, but is not limited to, lane markings, number and distribution of lanes, turning rules, special lanes, and road markings for the current road segment. Traffic snapshot data may include, but is not limited to, traffic congestion information, traffic accident information, road closure status, traffic flow and speed, and vehicle density for a specified time period (e.g., 30 seconds) before a network outage. Estimated Time of Arrival (ETA) indicates the estimated time the vehicle will reach its destination. Battery information indicates information related to the battery level of new energy vehicles, including, but not limited to, the location of charging stations along the route, remaining battery power, predicted waypoints, estimated distance to the destination, and estimated battery level.

[0085] The closed-loop parameters for assisted driving may include, but are not limited to, real-time environment adaptation parameters, vehicle control calibration parameters, and decision logic adaptation parameters. Real-time environment adaptation parameters may include, but are not limited to, lane curvature compensation values ​​for the current road segment and dynamic thresholds for following distance. Vehicle control calibration parameters may include, but are not limited to, steering angle correction coefficients and throttle / brake response sensitivity. Decision logic adaptation parameters may include, but are not limited to, lane change timing judgment thresholds and roundabout priority decision weights.

[0086] It should be noted that the difference between the first online data and the second online data is that the first online data can be the online data obtained after detecting an abnormal network state and then recovering under specified conditions. The specified conditions can be the conditions corresponding to the decision-making first mode. The second online data can be the online data obtained when the network was in a normal connection state before the network anomaly was detected.

[0087] In step S530, data validity is marked by timestamps. When marking data validity times, the data type and usage scenario can be considered. Different validity timestamps can be set for different types of data; furthermore, for the same type of data, corresponding validity timestamps can be marked according to different usage scenarios. For example, when marking validity, data with "road conditions longer than 5 minutes" is marked as expired because road conditions are real-time, and information older than 5 minutes has low reference value for current driving decisions. For real-time road condition snapshots, "road conditions 30 seconds before the interruption" can be specifically referred to. This data needs a shorter validity period to accurately connect road conditions before and after the interruption, avoiding decision-making biases caused by time differences. Both can have different validity times set according to the data's application scenario, and different validity time values ​​can be set according to specific needs in a particular scenario.

[0088] If an abnormal network condition is detected during the continuous acquisition of second online data, then... Figure 5 In steps S540-S570, offline cached data is determined from the second online data. In step S540, a network anomaly is triggered. When an abnormal network state is detected, step S550 is executed to dynamically trim the acquired second online data according to the vehicle location to obtain trimmed cached data.

[0089] The dynamic pruning process includes: when a network anomaly is detected, firstly, the vehicle's position at the time of the network interruption (i.e., the first vehicle position) and the full cached second online data before the interruption are obtained; then, the cumulative offline driving distance is calculated by combining the vehicle's real-time speed and the driving time after the network anomaly; finally, data pruning is performed according to the rule of retaining only the second online data within a specified distance range (such as an offline driving distance threshold) from the current position. For example, redundant data of already traversed road segments are removed, core information such as lane lines, road condition snapshots, and charging station locations within a specified distance range ahead is retained, while unnecessary data outside the specified distance range is discarded, resulting in pruned cached data that is compact and adapted to the current driving scenario.

[0090] In step S560, expired data is marked. The timestamps of the clipped cache data are compared with their corresponding valid time values. Data whose timestamps exceed the valid time values ​​are marked as expired data, and the marked expired data is deleted. Clipped cache data that is still within its validity period is selected as offline cache data. The offline cache data is compact, easy to store, and all data is within its validity period, and is adapted to the current driving scenario. In step S570, the offline cache data is output. The subsequent offline cache data can serve as data support for generating vehicle driving schemes for different driving models.

[0091] Continue to refer to Figure 4 ,exist Figure 1 Based on the driving mode determination method shown, Figure 1 As shown in step S130, the second mode can also be determined through the following steps. When it is determined in step S420 that the vehicle's offline driving distance is less than or equal to the offline driving distance threshold, in step S430, the relationship between the network recovery probability and the recovery probability threshold is further determined. In step S440, when the vehicle's offline driving distance is less than or equal to the offline driving distance threshold and the network recovery probability is less than the recovery probability threshold, in step S450, the driving mode is determined to be the second mode, i.e., the offline route optimization mode, and offline route optimization is initiated.

[0092] When the driving mode is in the second mode, the vehicle's driving plan can be generated through the following steps: Obtain offline cached data and directly optimize the vehicle's offline driving path using this data. Update the offline driving path using the offline cached data. For example, first, retrieve the traffic snapshot (including road congestion, speed limits, and intersection information) cached 30 seconds before the interruption. Then, combine this with the current vehicle location, speed, and preset destination. Use offline algorithms to remove outdated congested road sections from the cache, correct temporarily closed routes, and optimize and update the offline driving path to generate the optimal driving route suitable for the current scenario, which serves as the vehicle's driving plan in the second mode.

[0093] Offline driving routes can be the vehicle's travel path determined based on offline maps when the network is down. Offline maps can be pre-downloaded map data, allowing for location viewing and route planning even without a network connection. Furthermore, when driving mode is in the second mode, monitoring frequency can be increased to more promptly detect network recovery. By updating offline driving routes, it can be ensured that even if the offline driving distance threshold has not been reached, the route still meets the road conditions and battery level planning requirements before the network downtime, without relying on online data.

[0094] Continue to refer to Figure 4 ,exist Figure 1 Based on the driving mode determination method shown, Figure 1 As shown in step S130, the third mode can also be determined through the following steps. In step S420, it is determined whether the offline driving distance is less than or equal to the offline driving distance threshold. If the offline driving distance is greater than the offline driving distance threshold, the driving mode is determined to be the third mode, that is, the LCC downgrade mode is triggered, and the LCC downgrade is triggered. After the driving mode is determined to be the third mode, a warning message is sent to the user. The warning message sent to the user includes that the vehicle's driving mode has exited the assisted driving mode and switched to the third mode, so that the user can take over the vehicle in time and ensure driving safety.

[0095] In exemplary embodiments, warning prompts can be issued to the driver in various forms. For example, based on a head-up display (HUD) system, the warning prompt is projected in front of the driver's line of sight. For example, based on a voice prompt system, a warning prompt is issued to the driver via voice prompt to exit the assisted driving mode and switch to a third mode. For example, based on a haptic feedback system, a warning prompt is issued to the driver via seat vibration or other tactile means. This disclosure does not limit the specific form in which warning prompt information is issued to the driver.

[0096] When the driving mode is the third mode, the vehicle's driving plan can be generated through the following steps: obtain offline cache data, and in LCC downgrade mode, extract the core path information (such as the basic driving trajectory) from the offline cache data as the vehicle's driving path information; in addition, the offline cache data can be used to verify the feasibility of the extracted core path in LCC mode, such as whether it matches the basic path requirements of lane centering control, etc., and save the vehicle's driving path information so that the complete route can be quickly reconstructed after the network is restored.

[0097] Additionally, after the vehicle switches to the third driving mode, the execution status is fed back, such as the start time of the third mode and driving mode switching information. This driving mode switching information may include, but is not limited to, changes in driving mode and route switching success rate. By feeding back the execution status, dynamic decision-making basis can be provided for the vehicle to generate subsequent driving plans, specifically in two aspects: (1) Closed-loop adjustment decision. The effectiveness of the offline route optimization scheme is judged by the "route switching success rate" (if the success rate is low, the cached data needs to be called again to adjust the route). The smoothness of the vehicle control function connection is evaluated by the "LCC start time" (if the start timeout occurs, the backup control logic needs to be triggered) to avoid the disconnect between decision and actual execution.

[0098] (2) Risk warning and emergency response. If the feedback execution status is abnormal (such as route switching failure or LCC start delay), the decision subsystem can quickly identify the fault point and promptly initiate emergency measures (such as prompting the driver to take over or switching to a more stable offline control mode) to ensure driving safety.

[0099] The difference between the second and third modes lies in their functional objectives and triggering logic. The second mode focuses on the "route planning level," aiming to ensure route feasibility through offline cached data. The third mode focuses on the "vehicle control level," often triggered by sensor malfunctions or environmental perception failures. Its goal is to reduce the vehicle control level, such as downgrading from full-speed LCC to low-speed following, ensuring basic driving safety. This mode is not directly related to network status. When the driving mode is the second mode, the assisted driving mode is not immediately disengaged; instead, the offline path is optimized. When the driving mode is the third mode, the assisted driving mode is disengaged, and a higher-level driving mode downgrade is performed. All three driving modes require the use of offline cached data to generate the vehicle's driving plan, which is crucial for ensuring route continuity, switching stability, and downgrade safety when switching between different driving modes.

[0100] After determining the different driving modes of the vehicle, corresponding instructions can be generated according to the driving mode to control the vehicle to switch driving modes and generate driving plans for the vehicle in different driving modes. Figure 6 This is a flowchart illustrating the generation of driving plans based on different driving modes according to an exemplary embodiment of this disclosure. In step S610, a driving mode command is received. In step S620, the command type is determined. In step S630, first online data is loaded, and the HNOA parameters are updated. In step S640, lane-level navigation is maintained, and a successful execution feedback is provided. In step S650, advanced functions are disabled, and lane centering control is activated. In step S660, a voice prompt is given to the user, providing feedback on the LCC activation status.

[0101] Figure 7 This is a system architecture diagram illustrating an implementation of a driving mode determination method according to an exemplary embodiment of the present disclosure. The present disclosure can implement the above-described driving mode determination method through a driving mode determination system 700, which may include: a network monitoring subsystem 710, a location monitoring subsystem 720, a data caching subsystem 730, a driving mode decision subsystem 740, and a vehicle driving control subsystem 750.

[0102] Among them, the network monitoring subsystem 710 is used to determine the network status and recovery prediction probability based on the acquired network status data. For example, by collecting network status data such as network signal strength, latency, and packet loss rate, and combining it with historical network status data, the network recovery probability is predicted.

[0103] The location monitoring subsystem 720 is used to determine the vehicle's offline driving distance in response to an abnormal network state, and to determine whether the vehicle has entered the critical range of offline driving distance when the network is in an abnormal state.

[0104] The data caching subsystem 730 is used to acquire the offline cached data corresponding to the vehicle. When the network is in a normal connection state, the data caching subsystem 730 continuously caches the second online data (such as lane-level information, power planning, etc.), and when the network is in an abnormal state, it dynamically updates the offline cached data based on the second online data.

[0105] The mode decision subsystem 740 is used to determine the vehicle's current driving mode based on offline driving distance and recovery prediction probability. In this embodiment, there are three modes. The vehicle mode can be determined using a pre-built driving mode classification model or a pre-configured rule algorithm.

[0106] The pre-configured rule algorithm implementation process includes: taking an offline driving distance threshold of 3 kilometers and a recovery prediction probability threshold of 60% as an example, if the vehicle is in a "critical warning state" (distance ≤ 3 kilometers) and the recovery probability ≥ 60%, the driving mode is determined to be the first mode, and the rule of "prioritizing online route recovery and calling offline cached data to temporarily maintain navigation" is executed. When the first online data is received, the first online data and offline cached data are combined to maintain lane-level navigation; if the vehicle is in a "critical warning state" but the recovery probability < 60%, the rule of "starting offline route optimization and increasing network monitoring frequency" is executed; if it enters a "degradation trigger state" (distance > 3 kilometers), the rule of "immediately triggering LCC degradation and synchronously saving the current route data" is executed.

[0107] The driving control subsystem 750 is used to generate the driving route of the vehicle based on the vehicle's driving mode and offline cached data.

[0108] Figure 8 This is a timing diagram illustrating a driving mode determination method according to an exemplary embodiment of the present disclosure. The driving mode determination system 700 follows... Figure 8 The processing timing determines the vehicle's driving mode and generates a driving plan. When the network is in a normal connection state, the data caching subsystem 730 continuously caches the second online data, and the network monitoring subsystem 710 collects network status data and monitors the network status at the first monitoring frequency. When an abnormal network state is detected, the location monitoring subsystem 720 starts timing and distance calculation to obtain the vehicle's cumulative driving distance under abnormal network conditions, which is used as the offline driving distance. At the same time, the network monitoring subsystem 710 switches to high-frequency mode and collects network status data at the second monitoring frequency.

[0109] During vehicle operation, the location monitoring subsystem 720 continuously determines whether the vehicle's offline driving distance has entered a critical range, and simultaneously sends the monitored network status to the mode decision subsystem 740. The mode decision subsystem 740, combining the network recovery probability with the vehicle's offline driving distance, triggers route recovery or downgrade commands. If the vehicle recovers the network within the offline driving distance threshold, the driving control subsystem 750 loads the first online data to maintain normal navigation. Alternatively, if the vehicle is within the offline driving distance threshold but the network recovery probability has not reached the threshold, the driving control subsystem 750 initiates offline route optimization, optimizing the offline driving path based on offline cached data to generate a driving plan for the vehicle. If the vehicle does not recover the network within the offline driving distance threshold, the system downgrades to LCC (Local Control Center) and waits for network recovery before reconstructing the route.

[0110] Through data interaction and collaborative decision-making among various subsystems, route recovery can be achieved after network anomalies. On the one hand, by linking network monitoring and distance monitoring, route recovery can be triggered preferentially within the offline driving distance threshold, reducing the probability of system downgrade to LCC and maintaining the continuity of advanced driver assistance systems. On the other hand, when an abnormal network state is detected, key data is dynamically cached, and online data is quickly integrated after network recovery to avoid route gaps. By combining vehicle position, speed, and network recovery probability prediction, the frequency of network monitoring and driving mode strategies can be dynamically adjusted to improve adaptability to complex scenarios (such as tunnels and mountainous areas). When the vehicle's offline driving distance is within the threshold range but network recovery is not possible under abnormal network conditions, LCC downgrade preparation can be initiated in advance, reducing the impact of driving mode switching on safety through a smooth transition.

[0111] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the method disclosed herein.

[0112] Figure 9 This is a block diagram illustrating a driving mode determination device according to an exemplary embodiment of the present disclosure. The device of this embodiment can be applied to electronic devices.

[0113] like Figure 9 As shown, the driving mode determination device 900 may include: an offline distance determination unit 910, a network recovery prediction unit 920, and a driving mode determination unit 930.

[0114] The offline distance determination unit 910 is used to determine the offline driving distance of the vehicle in response to an abnormal network state; the network recovery prediction unit 920 is used to determine the network recovery probability by acquiring network status data; and the driving mode determination unit 930 is used to determine the driving mode of the vehicle based on the offline driving distance and the network recovery probability.

[0115] In some exemplary embodiments of this disclosure, the driving mode determination unit 930 includes a driving mode determination subunit; the driving mode determination subunit is used to input the offline driving distance and the network recovery probability into a pre-trained driving mode classification model to obtain the driving mode; the driving mode classification model is trained through the following steps: acquiring model training data, the model training data including model input features and driving mode classification results, the model input features including network recovery probability data and offline driving distance data; training an initial model using the model training data to obtain the driving mode classification model.

[0116] In some exemplary embodiments of this disclosure, the driving mode determination unit 930 includes at least a first mode determination unit, a cached data acquisition unit, a temporary navigation generation unit, an online data acquisition unit, and a first scheme generation unit; The first mode determination unit is configured to determine the driving mode as a first mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being greater than or equal to a recovery probability threshold; the cached data acquisition unit is configured to acquire offline cached data in response to the driving mode being the first mode; the temporary navigation generation unit is configured to generate a temporary navigation planning scheme for the vehicle based on the offline cached data; the online data acquisition unit is configured to acquire first online data in response to the network status being in a connected state; and the first scheme generation unit is configured to generate a driving scheme for the vehicle based on the first online data and the offline cached data.

[0117] In some exemplary embodiments of this disclosure, the driving mode determination unit 930 includes at least a second mode determination unit, a cached data acquisition unit, and an offline path optimization unit; the second mode determination unit is used to determine the driving mode as a second mode in response to the offline driving distance being less than or equal to an offline driving distance threshold and the network recovery probability being less than a recovery probability threshold; the cached data acquisition unit is used to acquire offline cached data in response to the driving mode being the second mode; the offline path optimization unit is used to update the offline driving path of the vehicle using the offline cached data, and generate a driving plan for the vehicle based on the updated offline driving path.

[0118] In some exemplary embodiments of this disclosure, the driving mode determination unit 930 includes at least a third mode determination unit, a warning notification unit, a cached data acquisition unit, and an information feedback unit; the third mode determination unit is used to determine the driving mode as a third mode in response to the offline driving distance being greater than an offline driving distance threshold; the warning notification unit is used to send a warning notification to the user in response to the driving mode being the third mode, the warning notification including that the vehicle has exited the assisted driving mode; the cached data acquisition unit is used to acquire offline cached data and determine the vehicle's driving path information based on the offline cached data; the information feedback unit is used to return the start time of the third mode and the switching information of the driving mode.

[0119] In some exemplary embodiments of this disclosure, the cached data acquisition unit is configured to acquire second online data, which is acquired when the network state is in a connected state; determine a timestamp corresponding to the second online data; in response to the network state being in the abnormal state, determine a valid time value corresponding to the second online data; and determine the offline cached data from the second online data based on the vehicle location, the valid time value, and the timestamp.

[0120] Figure 10 This is a functional block diagram of a vehicle according to an exemplary embodiment of the present disclosure. For example, vehicle 1000 can be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. Vehicle 1000 can be a driver-assisted vehicle, a semi-driver-assisted vehicle, or a driver-free vehicle.

[0121] Reference Figure 10 The vehicle 1000 may include various subsystems, such as an infotainment system 1010, a perception system 1020, a decision control system 1030, a drive system 1040, and a computing platform 1050. The vehicle 1000 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and component of the vehicle 1000 can be interconnected via wired or wireless means.

[0122] In some embodiments, the infotainment system 1010 may include a communication system, an entertainment system, and a navigation system, etc.

[0123] The perception system 1020 may include several types of sensors for sensing information about the environment surrounding the vehicle 1000. For example, the perception system 1020 may include a global positioning system (which may be a GPS system, a BeiDou system, or another positioning system), an inertial measurement unit (IMU), a lidar, a millimeter-wave radar, an ultrasonic radar, and a camera device.

[0124] The decision control system 1030 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0125] The drive system 1040 may include components that provide powered motion to the vehicle 1000. In one embodiment, the drive system 1040 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0126] Some or all of the functions of the vehicle 1000 are controlled by a computing platform 1050. The computing platform 1050 may include at least one processor 1051 and a memory 1052, the processor 1051 being able to execute instructions 1053 stored in the memory 1052.

[0127] The processor 1051 can be any conventional processor, such as a commercially available CPU. The processor may also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), systems on chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0128] The memory 1052 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0129] In addition to instruction 1053, memory 1052 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in memory 1052 can be used by computing platform 1050.

[0130] In this embodiment of the disclosure, the processor 1051 may execute instructions 1053 to complete all or part of the steps of the driving mode determination method described above.

[0131] In some embodiments of this disclosure, a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by the processor of a mobile terminal, enables the mobile terminal to perform all or part of the steps of the driving mode determination method described above.

[0132] In some embodiments of this disclosure, a computer program is provided, which, when executed by a processor, implements all or part of the steps of the driving mode determination method described above.

[0133] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0134] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for determining a driving mode, characterized in that, include: In response to an abnormal network status, determine the vehicle's offline driving distance; By acquiring network status data, the probability of network recovery can be determined. The driving mode of the vehicle is determined based on the offline driving distance and the network recovery probability.

2. The method according to claim 1, characterized in that, The response to an abnormal network status, determining the vehicle's offline driving distance, includes: In response to the network status being in the abnormal state, the vehicle's travel time and speed are obtained; The offline driving distance is determined based on the driving time and the driving speed.

3. The method according to claim 1, characterized in that, The step of determining the network recovery probability by acquiring network status data includes: Obtain the network status data; Obtain the historical network status data corresponding to the network status data; The network recovery probability is determined based on the network status data and the historical network status data.

4. The method according to claim 3, characterized in that, The acquisition of the network status data includes: In response to the network status being in a connected state, the network status data is acquired based on a first monitoring frequency; In response to the network status being in the abnormal state, network status data is acquired based on a second monitoring frequency, wherein the second monitoring frequency is higher than the first monitoring frequency.

5. The method according to claim 1, characterized in that, Determining the vehicle's driving mode based on the offline driving distance and the network recovery probability includes: The offline driving distance and the network recovery probability are input into a pre-trained driving mode classification model to obtain the driving mode; The driving mode classification model is trained through the following steps: Obtain model training data, which includes model input features and driving mode classification results. The model input features include network recovery probability data and offline driving distance data. The driving mode classification model is obtained by training the initial model using the model training data.

6. The method according to claim 1, characterized in that, Determining the vehicle's driving mode based on the offline driving distance and the network recovery probability includes: In response to the offline driving distance being less than or equal to the offline driving distance threshold and the network recovery probability being greater than or equal to the recovery probability threshold, the driving mode is determined as the first mode; In response to the driving mode being the first mode, offline cached data is obtained; A temporary navigation plan for the vehicle is generated based on offline cached data; In response to the network status being connected, the first online data is obtained; A driving plan for the vehicle is generated based on the first online data and the offline cached data.

7. The method according to claim 1, characterized in that, The step of determining the vehicle's driving mode based on the offline driving distance and the network recovery probability further includes: In response to the offline driving distance being less than or equal to the offline driving distance threshold and the network recovery probability being less than the recovery probability threshold, the driving mode is determined to be the second mode; In response to the driving mode being the second mode, offline cached data is obtained; The offline driving path of the vehicle is updated using the offline cached data, and a driving plan for the vehicle is generated based on the updated offline driving path.

8. The method according to claim 1, characterized in that, The step of determining the vehicle's driving mode based on the offline driving distance and the network recovery probability further includes: In response to the offline driving distance being greater than the offline driving distance threshold, the driving mode is determined to be the third mode; In response to the driving mode being the third mode, a warning message is sent to the user, the warning message including that the vehicle has exited the assisted driving mode; Obtain offline cached data, and determine the vehicle's driving path information based on the offline cached data; Return the start time of the third mode and the switching information of the driving mode.

9. The method according to any one of claims 6, 7 or 8, characterized in that, The process of obtaining offline cached data includes: Acquire second online data, which is online data acquired before the network status was detected to be in an abnormal state; Determine the timestamp marker corresponding to the second online data; In response to the network status being in the abnormal state, a valid time value corresponding to the second online data is determined; The offline cached data is determined from the second online data based on the vehicle location, the valid time value, and the timestamp.

10. The method according to claim 9, characterized in that, The second online data includes one or more of the following: lane-level navigation data, traffic snapshot data, estimated arrival time, battery information, and assisted driving closed-loop parameters.

11. The method according to claim 9, characterized in that, The step of determining the offline cached data from the second online data based on the vehicle location, the valid time value, and the timestamp includes: Based on the vehicle location, the valid time value, and the timestamp, the second online data is dynamically trimmed to obtain the offline cached data.

12. A driving mode determination device, characterized in that, include: The offline distance determination unit is used to determine the offline driving distance of the vehicle in response to an abnormal network state. The network recovery prediction unit is used to determine the probability of network recovery by acquiring network status data; A driving mode determination unit is used to determine the driving mode of the vehicle based on the offline driving distance and the network recovery probability.

13. The apparatus according to claim 12, characterized in that, The driving mode determination unit includes a driving mode determination subunit; The driving mode determination subunit is used to input the offline driving distance and the network recovery probability into a pre-trained driving mode classification model to obtain the driving mode; The driving mode classification model is trained through the following steps: Obtain model training data, which includes model input features and driving mode classification results. The model input features include network recovery probability data and offline driving distance data. The driving mode classification model is obtained by training the initial model using the model training data.

14. The apparatus according to claim 12, characterized in that, The driving mode determination unit includes at least a first mode determination unit, a cached data acquisition unit, a temporary navigation generation unit, an online data acquisition unit, and a first scheme generation unit; The first mode determination unit is configured to determine the driving mode as the first mode in response to the offline driving distance being less than or equal to the offline driving distance threshold and the network recovery probability being greater than or equal to the recovery probability threshold. A cached data acquisition unit is configured to acquire offline cached data in response to the driving mode being the first mode. A temporary navigation generation unit is used to generate a temporary navigation planning scheme for the vehicle based on offline cached data; An online data acquisition unit is configured to acquire first online data in response to the network status being in a connected state. The first scheme generation unit is used to generate a driving scheme for the vehicle based on the first online data and the offline cached data.

15. The apparatus according to claim 12, characterized in that, The driving mode determination unit includes at least a second mode determination unit, a cached data acquisition unit, and an offline path optimization unit; The second mode determination unit is configured to determine the driving mode as the second mode in response to the offline driving distance being less than or equal to the offline driving distance threshold and the network recovery probability being less than the recovery probability threshold. The cached data acquisition unit is used to acquire offline cached data in response to the driving mode being the second mode; The offline path optimization unit is used to update the offline driving path of the vehicle using the offline cached data, and generate a driving plan for the vehicle based on the updated offline driving path.

16. The apparatus according to claim 12, characterized in that, The driving mode determination unit includes at least a third mode determination unit, a warning prompt unit, a cached data acquisition unit, and an information feedback unit; The third mode determination unit is used to determine the driving mode as the third mode in response to the offline driving distance being greater than the offline driving distance threshold. The warning unit is used to send a warning message to the user in response to the driving mode being the third mode. The warning message includes that the vehicle has exited the assisted driving mode. The cached data acquisition unit is used to acquire offline cached data and determine the vehicle's driving path information based on the offline cached data; The information feedback unit is used to return the start time of the third mode and the switching information of the driving mode.

17. The apparatus according to any one of claims 14, 15 or 16, characterized in that, The cache data acquisition unit includes a cache data acquisition subunit; The cached data acquisition subunit is used to acquire second online data, which is online data acquired before the network state was detected to be in an abnormal state; Determine the timestamp marker corresponding to the second online data; In response to the network status being in the abnormal state, a valid time value corresponding to the second online data is determined; The offline cached data is determined from the second online data based on the vehicle location, the valid time value, and the timestamp.

18. A vehicle, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured as follows: The steps of implementing the driving mode determination method according to any one of claims 1 to 11.

19. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the mobile terminal, the mobile terminal is able to perform the steps of the driving mode determination method according to any one of claims 1 to 11.

20. A computer program, characterized in that, It includes a computer program that, when executed by a processor, implements the driving mode determination method as described in any one of claims 1 to 11.

Images