Driving training turning auxiliary method and device

Abstract

The invention relates to a driving training turning assisting method and device. The method comprises the steps that driving data and environment data of a vehicle are obtained; determining a current driving state according to the driving data; when the current driving state is a turning state, the driving condition is predicted according to the driving data, and a movement track is obtained; when an obstacle is detected in the environment data and the obstacle is on the motion trail, determining an auxiliary intervention level according to the driving data and the data of the obstacle, and assisting the vehicle through an auxiliary intervention scheme corresponding to the auxiliary intervention level; the problem that an existing driving training system cannot effectively predict and process the turning collision risk is solved by acquiring vehicle driving data and environment data in real time, predicting the motion trail and dynamically judging the obstacle risk in the turning state and combining with an auxiliary intervention scheme corresponding to the auxiliary intervention level; the method has the advantages of improving the accuracy of obstacle collision risk prediction and effectiveness of auxiliary measures in the turning process.

Description

Technical Field

[0001] This invention relates to the field of vehicle obstacle avoidance technology, and in particular to a method and device for assisting driver training in turning. Background Technology

[0002] With the development of technology, more and more intelligent systems and modules are being applied to the driver training field. The application of robot AI instructors not only meets students' higher requirements for convenience and safety, but also frees up more instructors from the passenger seat. One instructor can serve more students at the same time, improving driver training efficiency. The continuous intelligentization of traditional driving schools is an inevitable trend.

[0003] During driver training, novice learners, due to their lack of skill, need robot AI instructors to guide them when driving on winding roads. However, the existing robot AI instructors only partially replace the fixed training content of human instructors in the second subject, and cannot predict the potential collision risks during the third subject, which may cause the vehicle to fail to pass obstacles smoothly or to collide when turning.

[0004] Therefore, there is an urgent need to propose a driving training turning assistance method and device to solve the technical problem that the existing technology of robot AI coach only partially replaces the fixed training content of human instructors in the second subject, and cannot predict the potential collision risks in the turning process in the third subject, resulting in the vehicle being unable to pass through obstacles smoothly or causing a collision when turning. Summary of the Invention

[0005] In view of this, it is necessary to provide a driving training turning assistance method and device to solve the technical problem that the existing technology of robot AI coach only partially replaces the fixed training content of human instructors in the second subject, and cannot predict the potential collision risks in the turning process in the third subject, resulting in the vehicle being unable to pass through obstacles smoothly or causing a collision when turning.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a driving training turning assistance method, comprising: Acquire vehicle driving data and environmental data; Based on the driving data, determine the current driving status; When the current driving state is a turning state, the driving situation is predicted based on the driving data to obtain the motion trajectory; When an obstacle is detected in the environmental data and the obstacle is on the trajectory, the level of assistance intervention is determined based on the driving data and the obstacle data, and the vehicle is assisted through the assistance intervention scheme corresponding to the level of assistance intervention.

[0007] In one possible implementation, the driving data includes yaw rate, lateral acceleration, and steering wheel angle; determining the current driving state based on the driving data includes: The mean values ​​of the yaw rate and the mean value of the lateral acceleration of the vehicle within a preset time period are calculated to obtain the mean yaw rate and the mean lateral acceleration. When the average yaw angle is greater than the preset yaw angle, the average lateral yaw angle is greater than the preset yaw angle, and the steering wheel angle is greater than the preset steering angle, the current driving state is confirmed to be a turning state.

[0008] In one possible implementation, the driving data further includes driving speed, steering ratio, and vehicle wheelbase; the step of predicting the driving situation based on the driving data to obtain the motion trajectory includes: Calculate the tangent of the ratio of the steering wheel angle to the instantaneous turning radius, and use the ratio of the vehicle wheelbase to the tangent as the instantaneous turning radius; The ratio of the driving speed to the instantaneous turning radius is taken as the angular velocity; The trajectory is predicted based on the angular velocity and the instantaneous turning radius after the predicted time, thus obtaining the motion trajectory.

[0009] In one possible implementation, when an obstacle is detected in the environmental data and the obstacle is on the trajectory, the level of assistance intervention is determined based on the driving data and the obstacle data, including: When the obstacle is detected in the environmental data, the coordinate information of the obstacle is determined based on the environmental data; The coordinate information is converted to a coordinate system with the center of the rear wheel of the vehicle as the origin to obtain the coordinates of the obstacle; The predicted position of the obstacle is determined based on the obstacle coordinates and the instantaneous turning radius, and the predicted position of the trajectory is determined based on the instantaneous turning radius and the preset safety margin. When the predicted position of the obstacle is less than or equal to the predicted position of the trajectory, the level of assistance intervention is determined based on the driving data and the coordinates of the obstacle.

[0010] In one possible implementation, determining the level of assistance intervention based on the driving data and the obstacle coordinates includes: The impact time is determined based on the driving speed and the coordinates of the obstacle; The level of auxiliary intervention is determined based on the impact time.

[0011] In one possible implementation, determining the level of auxiliary intervention based on the impact time includes: When the impact time is greater than the first threshold, the auxiliary intervention level is confirmed to be the non-intervention level; When the impact time is less than or equal to the first threshold and the impact time is greater than the second threshold, the auxiliary intervention level is confirmed to be the light braking level; the first threshold is greater than the second threshold. When the impact time is less than or equal to the second threshold, the level of auxiliary intervention is confirmed as the emergency braking level.

[0012] In one possible implementation, assisting the vehicle through the assistance intervention scheme corresponding to the assistance intervention level includes: When the level of assistance intervention is the level of non-intervention, the vehicle is not controlled; When the level of assistance intervention is the light braking level, a prompt message is output, and a first braking force is obtained based on the driving speed, the current time, and the impact time. The vehicle is controlled based on a first percentage of the first braking force. When the level of assistance intervention is the emergency braking level, a second braking force is obtained based on the driving speed, the current time, and the impact time, and the vehicle is controlled based on a second percentage of the second braking force; the first percentage is less than the second percentage.

[0013] In one possible implementation, multiple sensors on the vehicle are respectively positioned in different areas; before determining the level of assistance intervention based on the driving data and the obstacle coordinates, the method further includes: When it is determined that the vehicle can safely pass the obstacle based on the steering wheel angle, the radar sensitivity coefficient corresponding to each area is set according to the distance from each probe to the center point of the rear wheel of the vehicle. The target coordinates are obtained by adjusting the obstacle coordinates based on the radar sensitivity coefficient. The determination of whether the obstacle is on the trajectory is based on the instantaneous turning radius, the preset safety margin, and the target coordinates.

[0014] In one possible implementation, determining whether the obstacle is on the trajectory based on the instantaneous turning radius, the preset safety margin, and the target coordinates includes: The product of the radar sensitivity coefficient of the region corresponding to the detected obstacle and the preset safety margin is used as the target safety margin. The predicted position of the target obstacle is determined based on the target coordinates and the instantaneous turning radius, and the predicted position of the target trajectory is determined based on the instantaneous turning radius and the target safety margin. When the predicted position of the target obstacle is less than or equal to the predicted position of the target trajectory, it is confirmed that the obstacle is on the movement trajectory.

[0015] Secondly, the present invention also provides a driver training turning assistance device, comprising: The data acquisition module is used to acquire vehicle driving data and environmental data; The type confirmation module is used to determine the current driving status based on the driving data; The trajectory confirmation module is used to predict the driving situation based on the driving data and obtain the motion trajectory when the current driving state is a turning state; The mode confirmation module is used to determine the level of assistance intervention based on the driving data and the data of the obstacle when an obstacle is detected in the environmental data and the obstacle is on the movement trajectory, and to assist the vehicle through the assistance intervention scheme corresponding to the level of assistance intervention.

[0016] The beneficial effects of this invention are: acquiring vehicle driving data and environmental data; determining the current driving state based on the driving data; when the current driving state is a turning state, predicting the driving situation based on the driving data to obtain the motion trajectory; when an obstacle is detected in the environmental data and the obstacle is on the motion trajectory, confirming the assistance intervention level based on the driving data and obstacle data, and assisting the vehicle through the assistance intervention scheme corresponding to the assistance intervention level; by acquiring vehicle driving data and environmental data in real time, predicting the motion trajectory and dynamically judging obstacle risk in the turning state, and combining the assistance intervention scheme corresponding to the assistance intervention level, this invention solves the problem that existing driver training systems cannot effectively predict and handle turning collision risks, and has the advantages of improving the accuracy of obstacle collision risk prediction and the effectiveness of assistance measures during turning. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of an embodiment of the driver training turning assistance method provided by the present invention; Figure 2 A schematic diagram of an embodiment of the vehicle turning safety assistance system provided by the present invention; Figure 3 For the present invention Figure 1 A schematic flowchart of an embodiment of step S103; Figure 4 For the present invention Figure 1 A schematic flowchart of an embodiment of step S104; Figure 5 For the present invention Figure 4 A schematic diagram of an embodiment prior to step S405; Figure 6This is a schematic diagram of an embodiment of the driver training turning assistance device provided by the present invention. Detailed Implementation

[0018] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0019] like Figure 1 As shown, a specific embodiment of the present invention discloses a driver training turning assistance method, comprising: S101. Obtain vehicle driving data and environmental data.

[0020] In this embodiment, multiple sensors can be installed on the vehicle, positioned at different locations to collect different data from different angles. Driving data refers to dynamic parameters reflecting the vehicle's motion state, specifically data such as yaw rate, lateral acceleration, and steering wheel angle collected by onboard sensors, used to determine whether the vehicle is turning. Environmental data refers to the perception information of obstacles around the vehicle, specifically data such as obstacle position and speed collected by cameras, millimeter-wave radar, or lidar, used to identify potential collision risks.

[0021] S102. Determine the current driving status based on the driving data.

[0022] The current driving state can include turning, straight driving, reversing into a parking space, etc. This embodiment of the invention mainly describes the control of the vehicle in the turning state, and other aspects related to this embodiment of the invention are not limited here.

[0023] S103. When the current driving state is turning, predict the driving situation based on the driving data to obtain the motion trajectory.

[0024] Among them, the turning state refers to the driving phase in which the vehicle is performing a turning operation. This can be determined by analyzing whether the average yaw rate over a preset time period exceeds a threshold, which triggers the trajectory prediction function. The motion trajectory refers to the predicted driving path of the vehicle over a future period of time. This can be achieved by establishing a vehicle kinematic model to calculate the instantaneous turning radius and angular velocity, which is used to assess whether an obstacle is in a danger zone.

[0025] S104. When an obstacle is detected in the environmental data and the obstacle is on the trajectory, the level of assistance intervention is determined based on the driving data and the obstacle data, and the vehicle is assisted through the assistance intervention plan corresponding to the level of assistance intervention.

[0026] Among them, the auxiliary intervention level refers to the different control strategies adopted according to the collision risk level. Specifically, it can be determined by calculating the relative position of the obstacle and the vehicle and the collision time, and is used to achieve graded braking intervention.

[0027] Specifically, this method first acquires real-time vehicle motion parameters, including yaw rate, lateral acceleration, and steering wheel angle, using onboard sensors. When these parameters consistently exceed set thresholds, the vehicle is determined to be entering a turning state. Subsequently, the instantaneous turning radius is calculated based on parameters such as the vehicle's wheelbase and steering ratio, and the future trajectory is predicted by combining this with the current vehicle speed. Simultaneously, environmental perception equipment detects surrounding obstacles, converts their coordinates to the vehicle's coordinate system, and compares them with the predicted trajectory. If an obstacle is within the trajectory range, the collision time is further calculated, and the appropriate level of intervention—no intervention, light braking, or emergency braking—is selected based on the time margin. For example, if the collision time is less than a set threshold, emergency braking can be automatically applied to avoid an accident.

[0028] The embodiments of the present invention can be applied to vehicle turning safety assistance systems, such as... Figure 2 As shown, the vehicle turning safety assistance system may include a main control module, a safety control module, a vehicle signal acquisition module, a throttle control module, and an obstacle detection module. These modules are connected to the main control module via bus communication. The main control module, based on the configured vehicle model, obtains the positions of the front and rear of the vehicle and the various sensors in the obstacle detection module based on the rear wheel center point. By uploading real-time distance information to surrounding obstacles through the obstacle detection module, the system can determine the environmental data of the student vehicle. The vehicle signal acquisition module acquires vehicle driving data, such as speed and steering wheel angle signals. The main control module receives, processes, and feeds back the data collected by the vehicle signal acquisition module and the obstacle detection module. The safety control module then confirms the level of assistance intervention and controls the brake and clutch pedals of the manual transmission vehicle based on the intervention level. The throttle control module controls the vehicle's throttle signal, thus providing assistance to the vehicle.

[0029] Compared with existing technologies, this embodiment provides the following advantages: acquiring vehicle driving data and environmental data; determining the current driving state based on the driving data; predicting the driving situation and obtaining the trajectory based on the driving data when the current driving state is turning; when an obstacle is detected in the environmental data and the obstacle is on the trajectory, confirming the assistance intervention level based on the driving data and obstacle data, and assisting the vehicle through the assistance intervention plan corresponding to the assistance intervention level; by acquiring vehicle driving data and environmental data in real time, predicting the trajectory and dynamically judging obstacle risk in the turning state, and combining the assistance intervention plan corresponding to the assistance intervention level, this embodiment solves the problem that existing driver training systems cannot effectively predict and handle turning collision risks, and has the advantages of improving the accuracy of obstacle collision risk prediction and the effectiveness of assistance measures during turning.

[0030] In some embodiments of the present invention, the driving data includes yaw rate, lateral acceleration, and steering wheel angle; step S102 includes: The mean values ​​of the vehicle's yaw rate and lateral acceleration over a preset time period are calculated to obtain the mean yaw rate and the mean lateral acceleration.

[0031] Among these, yaw rate refers to the angular velocity of the vehicle's rotation about its vertical axis, which can be measured in real time using a gyroscope sensor to characterize the vehicle's dynamic response characteristics during steering. Lateral acceleration refers to the acceleration component of the vehicle in the lateral direction, which can be collected using an accelerometer to reflect the centrifugal force generated during turning. Steering wheel angle refers to the rotation angle of the steering wheel operated by the driver, which can be detected using an angle encoder to directly characterize the driver's steering intention. Therefore, the yaw rate can be calculated over a preset time to obtain the average yaw angle, and the lateral acceleration can also be calculated over a preset time to obtain the average lateral acceleration.

[0032] When the average yaw angle is greater than the preset yaw angle, the average lateral yaw angle is greater than the preset yaw angle, and the steering wheel angle is greater than the preset steering angle, the current driving state is confirmed as a turning state.

[0033] The preset yaw angle, preset yaw angle, and preset steering angle can all be set according to actual conditions. This embodiment of the invention does not impose any restrictions on these settings. For example, the preset yaw angle can be set to 5° / s, and the preset steering angle can be set to 5 degrees, to establish a quantitative standard for judging the turning state. The judgment of the turning state is shown in formula (1): (1) In the formula, The mean of the yaw angle. The lateral mean. For steering wheel turning angle, This is the summation operator.

[0034] Specifically, by continuously collecting yaw rate and lateral acceleration data and calculating their moving averages within a preset time window, instantaneous noise interference can be effectively eliminated. When the average yaw rate exceeds a preset threshold, it indicates that the vehicle is in a continuous turning state; when the average lateral rate exceeds a threshold, it indicates that there is sufficient centrifugal force; and when the steering wheel angle exceeds a threshold, it confirms that the driver has a clear turning intention. The combined judgment of these three conditions can accurately distinguish between straight-line driving and turning driving states, avoiding misjudgment based on a single parameter. For example, when driving on bumpy roads, although there may be instantaneous lateral acceleration fluctuations, the system will not misjudge it as a turning state because the steering wheel angle does not reach the threshold.

[0035] Through the above technical solution, this application can accurately identify the vehicle's turning state, providing a reliable basis for subsequent trajectory prediction and collision warning. The multi-parameter fusion judgment mechanism effectively avoids false triggering caused by abnormal data from a single sensor, ensuring that the system only activates the assistance function in real turning scenarios. This improves the timeliness of safety protection while avoiding unnecessary system intervention that could affect the driving experience.

[0036] In some embodiments of the present invention, the driving data also includes driving speed, steering ratio, and vehicle wheelbase; such as Figure 3 As shown, step S103 includes: S301. Calculate the tangent of the ratio of steering wheel angle to instantaneous turning radius, and use the ratio of vehicle wheelbase to tangent as instantaneous turning radius.

[0037] The instantaneous turning radius refers to the turning radius corresponding to the front wheel steering angle during vehicle turning. Specifically, it can be calculated by dividing the vehicle's wheelbase by the tangent of the product of the steering wheel angle and the steering ratio. This parameter is used to quantify the path curvature during vehicle turning. The instantaneous turning radius is calculated as shown in formula (2): (2) In the formula, Instantaneous turning radius (unit: meters). This refers to the vehicle's wheelbase. For steering wheel turning angle, This is the steering gear ratio.

[0038] S302. The ratio of driving speed to instantaneous turning radius is taken as angular velocity.

[0039] Angular velocity refers to the rate at which a vehicle rotates around the center of a turn. It can be calculated by dividing the vehicle's speed by the instantaneous turning radius. This parameter describes the dynamic rate of change in the vehicle's steering. Angular velocity ω = Vehicle speed V / Instantaneous turning radius R.

[0040] S303. Based on the angular velocity and instantaneous turning radius, predict the trajectory after the prediction time to obtain the motion trajectory.

[0041] The trajectory after the predicted time refers to the vehicle's driving path within a specified time range in the future, calculated based on the current motion state. Specifically, this can be achieved by constructing the motion equation in the polar coordinate system using angular velocity and instantaneous turning radius. This parameter is used to predict the spatial area that the vehicle may occupy during the turning process. Based on the instantaneous turning radius R and angular velocity ω, the predicted trajectory within the next T seconds is an arc, and the trajectory equation can be expressed as shown in formulas (3) and (4): (3) (4) In the formula, t ∈ [0, T] is the time to be predicted.

[0042] Specifically, the vehicle wheelbase, as a fixed parameter, is input into the steering ratio calculation model along with the real-time collected steering wheel angle. The front wheel steering angle is derived using trigonometric functions, and then the instantaneous turning radius is calculated based on the wheelbase parameter. The ratio of the driving speed to the instantaneous turning radius generates angular velocity. This angular velocity, combined with a time variable, constructs a circular motion trajectory equation in polar coordinates, based on the instantaneous turning radius, thereby predicting the vehicle's trajectory curve over a future time period. For example, when the vehicle makes a left turn at a specific speed, the system can continuously calculate the instantaneous turning radius and angular velocity to update the predicted driving path of the vehicle in real time for the next three seconds.

[0043] Through the above technical solution, this application can accurately predict the trajectory of a vehicle in a turning state, providing a reliable spatial position benchmark for subsequent obstacle collision judgment, effectively solving the problem in the prior art that the obstacle cannot be identified in a timely manner due to trajectory prediction deviation, and improving the safety warning accuracy of steering operation during driver training.

[0044] In some embodiments of the present invention, such as Figure 4 As shown, step S104 includes: S401. When an obstacle is detected in the environmental data, determine the coordinate information of the obstacle based on the environmental data. Obstacles can be detected using vehicle-mounted radar or cameras, or multiple probes can be used to collect information, and then the collected data can be used to detect obstacle information.

[0045] S402. Transform the coordinate information into a coordinate system with the center of the rear wheel of the vehicle as the origin to obtain the coordinates of the obstacle; S403. Determine the predicted position of the obstacle based on the obstacle coordinates and the instantaneous turning radius, and determine the predicted position of the trajectory based on the instantaneous turning radius and the preset safety margin; S404. When the predicted position of the obstacle is less than or equal to the predicted position of the trajectory, the level of assistance intervention shall be determined based on the driving data and the coordinates of the obstacle.

[0046] Among them, the coordinate information transformation to the coordinate system with the rear wheel center of the vehicle as the origin means to convert the absolute position of the obstacle into the relative position with respect to the vehicle's motion reference point. Specifically, it can be achieved by using a coordinate transformation matrix or a geometric projection algorithm. For example, after obtaining the latitude and longitude data of the obstacle in the global coordinate system through the vehicle's onboard sensors, the coordinate translation calculation is performed in combination with the vehicle's real-time positioning information. The instantaneous turning radius refers to the radius of the arc trajectory formed during the vehicle's turning process. Specifically, it can be calculated by the vehicle's wheelbase, steering wheel angle, and steering transmission ratio. For example, the theoretical turning radius can be derived using the Ackermann steering geometry model. The preset safety margin refers to the safety buffer range reserved for the vehicle's driving trajectory. Specifically, it can be set to a fixed value or dynamically adjusted according to the vehicle speed. For example, it can be set to 0.5 meters at low speed and increased proportionally to the vehicle speed at high speed. The following conditions are used to determine whether the obstacle is on the predicted trajectory, as shown in formula (5): (5) In the formula, As a preset safety margin, the default is 0.5m. Here are the coordinates of the obstacle. The instantaneous turning radius is given. The left side of formula (5) is for determining the predicted position of the obstacle based on the obstacle coordinates and the instantaneous turning radius, and the right side is for determining the predicted position of the trajectory based on the instantaneous turning radius and the preset safety margin.

[0047] Specifically, when the vehicle radar or camera detects an obstacle, its three-dimensional coordinate data is first extracted. The global coordinates of the obstacle are converted into local coordinates with the center of the rear wheel of the vehicle as the origin through the coordinate transformation module, eliminating the error caused by the change in the vehicle's own position. Combined with the instantaneous turning radius calculated in real time, a virtual trajectory band containing a preset safety margin is generated in the motion trajectory prediction model. The judgment is made by geometric calculation, as shown in formula (5). The obstacle coordinates are judged to be within the range of the trajectory band by calculating the predicted position of the obstacle and the predicted position of the trajectory. If the predicted position of the obstacle is less than or equal to the predicted position of the trajectory, it is determined that there is a risk of collision. When it is determined that there is no risk, the brake and clutch pedals will not be controlled, and the trainee will move with the current operation. When it is determined that there is a risk, the auxiliary intervention level decision process is triggered, and control commands are generated by combining the relative position of the obstacle and the dynamic parameters of the vehicle.

[0048] Through the above technical solutions, the dual protection of coordinate reference transformation and safety margin significantly improves the adaptability of the auxiliary system to complex steering conditions and avoids unnecessary braking or collision risks caused by trajectory prediction deviations.

[0049] In some embodiments of the present invention, step S405 includes: Determine the time of impact based on the vehicle speed and the coordinates of the obstacle; Determine the level of auxiliary intervention based on the impact time.

[0050] The impact time refers to the estimated time interval between a vehicle and an obstacle collision. It can be calculated as the ratio of the obstacle's coordinates to the vehicle's speed. For example, the real-time distance between the obstacle and the vehicle can be calculated using the Euclidean distance formula, and then divided by the speed to obtain the time value. This parameter quantifies the urgency of the collision risk, providing a dynamic basis for selecting the level of auxiliary intervention. The level of auxiliary intervention refers to the vehicle control strategy adopted for different risk levels. Different intervention levels can be defined by preset time thresholds, such as setting a first threshold and a second threshold as the trigger boundaries for light braking and emergency braking. This feature enables risk-level response, avoiding over-intervention or delayed response caused by a single braking strategy.

[0051] Specifically, when an obstacle is detected on the vehicle's trajectory, the system calculates the relative distance between the vehicle and the obstacle in real time. Combining this with the current driving speed, it calculates the impact time (TTC) from the current time until the vehicle hits the obstacle, using a formula such as TTC = obstacle distance / vehicle speed. For example, if the driving speed is 10 meters per second and the obstacle distance is 50 meters, the impact time is 5 seconds. Based on the comparison between this calculation result and a preset threshold, the system automatically selects between no intervention, light braking, or emergency braking. For instance, if the impact time is greater than 3 seconds, normal driving is maintained; if it is between 1 and 3 seconds, partial braking is applied; and if it is less than 1 second, full braking is triggered.

[0052] Through the above technical solution, this application achieves dynamic collision risk assessment based on real-time motion status, and reduces interference with the driving process while ensuring safety through a graded braking strategy. For example, it only provides a warning when approaching an obstacle slowly over a long distance, and automatically triggers emergency braking when approaching it at high speed over a short distance, thereby balancing safety and driving experience in complex turning scenarios.

[0053] In some embodiments of the present invention, the level of auxiliary intervention is determined based on the impact time, including: When the impact time exceeds the first threshold, the level of auxiliary intervention is confirmed to be the non-intervention level; When the impact time is less than or equal to the first threshold and greater than the second threshold, the assisted intervention level is confirmed as the light braking level; the first threshold is greater than the second threshold. When the impact time is less than or equal to the second threshold, the assisted intervention level is confirmed as the emergency braking level.

[0054] Among them, the first threshold and the second threshold are time parameters set in advance for distinguishing different risk levels. For example, the first threshold can be 3 seconds and the second threshold can be 1 second, which are determined through experimental data or driving behavior models. The assisted intervention level refers to the control strategy triggered according to the risk assessment result. For example, the non-intervention level only maintains the monitoring state. The light braking level applies partial braking force on the premise of maintaining vehicle controllability, while the emergency braking level triggers the maximum braking force to avoid collisions.

[0055] Specifically, after detecting that the obstacle is located on the predicted trajectory, the impact time is divided into three intervals for judgment. When the impact time exceeds the first threshold (i.e., TTC>3s), it indicates that there is sufficient safety margin and the system does not actively intervene in the driving operation; when the impact time is between the first threshold and the second threshold (i.e., 1s<TTC ≤ 3s), the system reduces the vehicle speed in advance at the light braking level and sends a warning signal to the driver at the same time; when the impact time is lower than the second threshold (i.e., TTC ≤ 1s), the system immediately activates the emergency braking to avoid collisions. This hierarchical control strategy realizes a smooth transition from early warning to active intervention by setting double time thresholds.

[0056] In some embodiments of the present invention, the vehicle is assisted through the assisted intervention scheme corresponding to the assisted intervention level, including: When the assisted intervention level is the non-intervention level, the vehicle is not controlled. When the assisted intervention level is the light braking level, a prompt message is output, and the first braking force is obtained according to the driving speed, the current time, and the impact time, and the vehicle is controlled according to the first braking force of the first percentage. When the assisted intervention level is the emergency braking level, the second braking force is obtained according to the driving speed, the current time, and the impact time, and the vehicle is controlled according to the second braking force of the second percentage; the first percentage is less than the second percentage.

[0057] Braking force refers to the force required for vehicle braking, which can be calculated by multiplying the vehicle speed by the inverse of the impact time. For example, it can be calculated by dynamically adjusting the calculation parameters based on real-time speed data obtained from a speed sensor and the distance to the obstacle. This parameter is used to quantify the distribution of braking force. The first percentage refers to the proportional coefficient of braking force at the light braking level, which can be achieved using a fixed value or a dynamic adjustment strategy. For example, it can be set to 20%-50% of the total braking force, ensuring smooth braking to avoid loss of vehicle control. The second percentage refers to the proportional coefficient of braking force at the emergency braking level, which can also be achieved using a fixed value or a dynamic adjustment strategy. For example, it can be set to 80%-100% of the total braking force, ensuring rapid deceleration to avoid collision risks. Warning information refers to alert signals to the driver, which can be implemented through voice prompts, screen displays, or steering wheel vibration. For example, a "Prompt obstacle ahead" voice warning can be output through the vehicle's speakers, assisting the driver in taking timely countermeasures.

[0058] Specifically, if an obstacle is detected on the trajectory during a turn, the system calculates the impact time using the vehicle speed and obstacle coordinates, and selects the intervention level based on the impact time threshold. When the impact time is within the light braking level range, the system outputs voice or visual prompts and applies a first percentage of braking force (e.g., 20%-50% braking force), gradually reducing the vehicle speed to give the driver sufficient time to take over. When the impact time reaches the emergency braking level threshold, the system directly applies a second percentage of braking force (e.g., 80%-100% braking force), for example, triggering maximum braking force through the electronic stability program to force the vehicle to decelerate. The no-intervention level maintains the vehicle's original state, only monitoring data changes in the background to cultivate the learner's independent judgment ability. Through this graded braking strategy, excessive intervention in driving operations is avoided while proactively mitigating risks in emergency situations.

[0059] Through the aforementioned technical solution, this application enables differentiated braking control based on collision risk levels during cornering. In the light braking phase, prompts guide learners to actively correct their actions, while in emergency situations, high-intensity braking ensures safety, thus addressing both instructional guidance and safety requirements. For example, after receiving a prompt, learners can adjust the steering wheel angle to avoid a collision without relying on the system forcibly taking over, effectively improving the practical effectiveness of driving training.

[0060] In some embodiments of the present invention, such as Figure 5 As shown, multiple sensors on the vehicle are installed in different areas; before step S405, the following steps are also included: S501. When it is determined that the vehicle can safely pass through an obstacle based on the steering wheel angle, the radar sensitivity coefficient corresponding to each area is set according to the distance from each probe to the center point of the rear wheel of the vehicle.

[0061] In certain special scenarios, the main control module determines whether an obstacle falls on the vehicle's trajectory and can assess whether the vehicle can safely pass the obstacle based on the current steering wheel angle. Area division refers to logically segmenting the space around the vehicle according to the probe's detection range. This can be achieved using geometric mesh partitioning or fan-shaped partitioning based on the detection angle. By dividing the space into different areas, data collected by each probe can be processed specifically. The radar sensitivity coefficient is an adjustment parameter reflecting the accuracy of obstacle coordinate measurement. It can be implemented using a distance attenuation function or a preset proportional coefficient. The farther the probe is from the center of the vehicle's rear wheels, the lower its sensitivity coefficient, thus compensating for the increase in measurement error with distance.

[0062] Specifically, if the obstacle can be safely passed by the current steering wheel angle, the various probes of the obstacle detection module can be divided into regions, and different radar sensitivity coefficients can be set for different regions based on the distance of each probe from the center point of the rear wheel of the vehicle.

[0063] S502. Adjust the obstacle coordinates according to the radar sensitivity coefficient to obtain the target coordinates.

[0064] The target coordinates refer to the obstacle position data after the sensitivity coefficient correction. Specifically, it can be achieved by using coordinate scaling or offset compensation algorithms. By adjusting the coordinate values, the interference of probe detection error on trajectory prediction can be eliminated.

[0065] S503. Determine whether the obstacle is on the trajectory based on the instantaneous turning radius, preset safety margin, and target coordinates.

[0066] Specifically, during vehicle turning, if the system determines that the current turning angle allows safe passage through an obstacle, it activates a probe area optimization mechanism. First, the radar probes distributed around the vehicle are divided into front, rear, and left / right zones based on their installation location. For example, probes 0-2 meters from the rear wheel center are defined as the near-field zone, 2-5 meters as the mid-field zone, and above 5 meters as the far-field zone. Based on the distance differences between the probes in each zone and the vehicle's center of motion, gradient parameters with sensitivity coefficients of 1.2, 1.0, and 0.6 are set respectively. When a probe detects an obstacle, the system reads the sensitivity coefficient of the region to which that probe belongs and performs weighted processing on the original coordinate data. For example, the coordinate data of the far-field zone is multiplied by a coefficient of 0.6 for shrinkage correction. The corrected target coordinates are then superimposed with the vehicle's instantaneous turning radius and combined with a preset safety margin threshold to generate a more accurate obstacle trajectory interference judgment result.

[0067] In some embodiments of the present invention, step S505 includes: The target safety margin is the product of the radar sensitivity coefficient of the area corresponding to the detected obstacle and the preset safety margin. The predicted position of the target obstacle is determined based on the target coordinates and the instantaneous turning radius, and the predicted position of the target trajectory is determined based on the instantaneous turning radius and the target safety margin. When the predicted position of the target obstacle is less than or equal to the predicted position of the target trajectory, the obstacle is confirmed to be on the trajectory.

[0068] Among them, the target safety margin refers to the safety distance margin after adjustment in combination with the radar sensitivity coefficient. Specifically, it can be achieved through multiplication coefficients or superposition of offsets, and is used to dynamically adapt to the detection errors in different areas.

[0069] Specifically, to balance safety and teaching effectiveness, a radar sensitivity coefficient K (0.6-1.2) is introduced to dynamically adjust the safety margin, resulting in the target safety margin, which is adjusted as follows: δ _Target safety margin= K × δ _Preset safety margin. The calculation of the predicted position of the target obstacle and the predicted position of the target trajectory is shown on the left and right sides of formula (5). Formula (5) can be used to determine the magnitude of the predicted position of the target obstacle and the predicted position of the target trajectory.

[0070] Specifically, when an obstacle is detected, the preset safety margin is first adjusted based on the radar sensitivity coefficient of the area where the obstacle is located. For example, if the obstacle is located in an area close to the center of the rear wheel, where the radar sensitivity coefficient is high, the target safety margin can be adjusted to 80% of the preset value; if the obstacle is located in a more distant area, where the sensitivity coefficient is low, the target safety margin is adjusted to 120% of the preset value. Subsequently, a safe trajectory range is constructed based on the instantaneous turning radius and the target safety margin. If the obstacle's coordinates fall within this range, it is determined that it is on the trajectory.

[0071] In some specific implementations, the radar sensitivity coefficient can be achieved through a preset mapping table. For example, the coefficient for the area 0-2 meters from the center of the rear wheel can be set to 1.2, for the area 2-4 meters to 1.0, and for the area above 4 meters to 0.8. The target safety margin can be generated through multiplication operations. For example, multiplying the preset safety margin of 1.5 meters by the sensitivity coefficient of 1.2 for the corresponding area yields a target safety margin of 1.8 meters.

[0072] Through the above technical solution, this application solves the problem of obstacle misjudgment caused by uneven sensor error distribution in the prior art. By adjusting the regional sensitivity, the setting of the safety margin is negatively correlated with the detection accuracy, which avoids excessive intervention while ensuring safety and improves the judgment accuracy and adaptability of the turning assist system.

[0073] To better implement the driver training turning assistance method in the embodiments of the present invention, based on the driver training turning assistance method, the embodiments of the present invention also provide a driver training turning assistance device, such as... Figure 6 As shown, the driver training turning assist device 600 includes: The data acquisition module 601 is used to acquire vehicle driving data and environmental data; The type confirmation module 602 is used to determine the current driving status based on driving data; The trajectory confirmation module 603 is used to predict the driving situation based on driving data and obtain the motion trajectory when the current driving state is turning. The pattern confirmation module 604 is used to confirm the level of assistance intervention based on the driving data and the obstacle data when an obstacle is detected in the environmental data and the obstacle is on the movement trajectory, and to assist the vehicle through the assistance intervention plan corresponding to the level of assistance intervention.

[0074] The driver training turning assistance device 600 provided in the above embodiments can realize the technical solutions described in the above driver training turning assistance method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above driver training turning assistance method embodiments, and will not be repeated here.

[0075] The driving training turning assistance method and device provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for assisting driving instructors in turning, characterized in that, include: Acquire vehicle driving data and environmental data; Based on the driving data, determine the current driving status; When the current driving state is a turning state, the driving situation is predicted based on the driving data to obtain the motion trajectory; When an obstacle is detected in the environmental data and the obstacle is on the trajectory, the level of assistance intervention is determined based on the driving data and the obstacle data, and the vehicle is assisted through the assistance intervention scheme corresponding to the level of assistance intervention.

2. The driver training turning assistance method according to claim 1, characterized in that, The driving data includes yaw rate, lateral acceleration, and steering wheel angle; determining the current driving state based on the driving data includes: The mean values ​​of the yaw rate and the mean value of the lateral acceleration of the vehicle within a preset time period are calculated to obtain the mean yaw rate and the mean lateral acceleration. When the average yaw angle is greater than the preset yaw angle, the average lateral yaw angle is greater than the preset yaw angle, and the steering wheel angle is greater than the preset steering angle, the current driving state is confirmed to be a turning state.

3. The driver training turning assistance method according to claim 2, characterized in that, The driving data also includes driving speed, steering ratio, and vehicle wheelbase; the step of predicting the driving situation based on the driving data to obtain the motion trajectory includes: Calculate the tangent of the ratio of the steering wheel angle to the instantaneous turning radius, and use the ratio of the vehicle wheelbase to the tangent as the instantaneous turning radius; The ratio of the driving speed to the instantaneous turning radius is taken as the angular velocity; The trajectory is predicted based on the angular velocity and the instantaneous turning radius after the predicted time, thus obtaining the motion trajectory.

4. The driver training turning assistance method according to claim 3, characterized in that, When an obstacle is detected in the environmental data and the obstacle is on the trajectory, the level of assistance intervention is determined based on the driving data and the obstacle data, including: When the obstacle is detected in the environmental data, the coordinate information of the obstacle is determined based on the environmental data; The coordinate information is converted to a coordinate system with the center of the rear wheel of the vehicle as the origin to obtain the coordinates of the obstacle; The predicted position of the obstacle is determined based on the obstacle coordinates and the instantaneous turning radius, and the predicted position of the trajectory is determined based on the instantaneous turning radius and the preset safety margin. When the predicted position of the obstacle is less than or equal to the predicted position of the trajectory, the level of assistance intervention is determined based on the driving data and the coordinates of the obstacle.

5. The driver training turning assistance method according to claim 4, characterized in that, The step of determining the level of assistance intervention based on the driving data and the obstacle coordinates includes: The impact time is determined based on the driving speed and the coordinates of the obstacle; The level of auxiliary intervention is determined based on the impact time.

6. The driver training turning assistance method according to claim 5, characterized in that, The step of determining the level of auxiliary intervention based on the impact time includes: When the impact time is greater than the first threshold, the auxiliary intervention level is confirmed to be the non-intervention level; When the impact time is less than or equal to the first threshold and the impact time is greater than the second threshold, the auxiliary intervention level is confirmed to be the light braking level; the first threshold is greater than the second threshold. When the impact time is less than or equal to the second threshold, the level of auxiliary intervention is confirmed as the emergency braking level.

7. The driver training turning assistance method according to claim 6, characterized in that, The method of assisting the vehicle through the assistance intervention scheme corresponding to the assistance intervention level includes: When the level of assistance intervention is the level of non-intervention, the vehicle is not controlled; When the level of assistance intervention is the light braking level, a prompt message is output, and a first braking force is obtained based on the driving speed, the current time, and the impact time. The vehicle is controlled based on a first percentage of the first braking force. When the level of assistance intervention is the emergency braking level, a second braking force is obtained based on the driving speed, the current time, and the impact time, and the vehicle is controlled based on a second percentage of the second braking force; the first percentage is less than the second percentage.

8. The driver training turning assistance method according to claim 4, characterized in that, The multiple sensors on the vehicle are respectively set in different areas; before confirming the level of assistance intervention based on the driving data and the obstacle coordinates, the process also includes: When it is determined that the vehicle can safely pass the obstacle based on the steering wheel angle, the radar sensitivity coefficient corresponding to each area is set according to the distance from each probe to the center point of the rear wheel of the vehicle. The target coordinates are obtained by adjusting the obstacle coordinates based on the radar sensitivity coefficient. The determination of whether the obstacle is on the trajectory is based on the instantaneous turning radius, the preset safety margin, and the target coordinates.

9. The driver training turning assistance method according to claim 8, characterized in that, The step of determining whether the obstacle is on the trajectory based on the instantaneous turning radius, the preset safety margin, and the target coordinates includes: The product of the radar sensitivity coefficient of the region corresponding to the detected obstacle and the preset safety margin is used as the target safety margin. The predicted position of the target obstacle is determined based on the target coordinates and the instantaneous turning radius, and the predicted position of the target trajectory is determined based on the instantaneous turning radius and the target safety margin. When the predicted position of the target obstacle is less than or equal to the predicted position of the target trajectory, it is confirmed that the obstacle is on the movement trajectory.

10. A driver training turning assist device, characterized in that, include: The data acquisition module is used to acquire vehicle driving data and environmental data; The type confirmation module is used to determine the current driving status based on the driving data; The trajectory confirmation module is used to predict the driving situation based on the driving data and obtain the motion trajectory when the current driving state is a turning state; The mode confirmation module is used to determine the level of assistance intervention based on the driving data and the data of the obstacle when an obstacle is detected in the environmental data and the obstacle is on the movement trajectory, and to assist the vehicle through the assistance intervention scheme corresponding to the level of assistance intervention.

Images