Adaptive lane keeping assist method and device for assisted driving

Abstract

The present application discloses an adaptive lane keeping assist method and device for assisted driving. A method and device for generating an action strategy for controlling an autonomous vehicle, wherein when the method and device are in operation, a vehicle sensor detects the distance from a host vehicle to a lane edge, determines the state of a vehicle system, a processor calculates a risk index in response to the state of the vehicle system, a processor calculates an intervention threshold in response to the risk index, a processor generates a vehicle path in response to a condition that the distance from the host vehicle to the lane edge is less than the intervention threshold, and a vehicle controller controls the vehicle in response to the vehicle path.

Description

Technical Field

[0001] The present disclosure generally relates to programming of autonomous motor vehicle control systems. More specifically, aspects of the present disclosure relate to a path shaping system, method, and apparatus for a lane keeping assist algorithm to enable the algorithm to adapt to the driver's preferences as well as the driver's state, vehicle state, and environmental conditions based on a calculated risk index formula. Background Art

[0002] The operation of modern vehicles is increasingly automated, i.e., driving control is achieved with less and less driver intervention. Vehicle automation is divided into a number of numerical levels from zero to five, where level zero corresponds to no automation with pure manual control and level five corresponds to pure automation with no manual control. Various automated driving assistance systems such as cruise control, adaptive cruise control and parking assistance systems correspond to lower automation levels, while truly "driverless" vehicles correspond to higher automation levels.

[0003] Automated driving assistance systems, such as lane keeping assist systems, typically determine a vehicle path in response to detection of obstacles, lane markings, road paths, and other external factors. These characteristic behaviors are often inconsistent with the expectations and personal preferences of vehicle occupants. In some cases, the vehicle's behavior under different driving conditions may be perceived as inconsistent with the expectations of the vehicle occupants or confusing. It is desirable to provide a method for vehicle path shaping during automated driving assistance operations that meets the expectations of vehicle occupants to improve occupant comfort and reduce vehicle occupant concerns.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0005] An autonomous vehicle control system training system and related control logic for implementing autonomous vehicle control, a method for manufacturing and operating such a system, and a motor vehicle equipped with an on-board control system are disclosed herein. By way of example and not limitation, a motor vehicle with an on-board vehicle control learning and control system for performing a lane keeping assist function in response to a changing risk level and a changing intervention threshold is presented.

[0006] According to one aspect of the present disclosure, a device is provided, which includes: a sensor configured to detect a distance from a host vehicle to an edge of a lane; a vehicle system configured to provide a status of the vehicle system; a processor configured to determine a risk index in response to the status of the vehicle system, calculate an intervention threshold in response to the risk index, and generate a vehicle path in response to a condition that the distance from the host vehicle to the edge of the lane is less than the intervention threshold; and a vehicle controller for controlling the vehicle in response to the vehicle path.

[0007] According to another aspect of the present disclosure, the state of the vehicle system is a driver's engagement level determined by a driver monitoring system.

[0008] According to another aspect of the present disclosure, the distance between the intervention threshold and the lane edge increases in proportion to the risk index.

[0009] According to another aspect of the present disclosure, the risk index is determined in response to at least one of pavement material, shoulder type, and road geometry.

[0010] According to another aspect of the present disclosure, the state of the vehicle system is a driver engagement level determined in response to a steering intervention by the driver.

[0011] According to another aspect of the present disclosure, the processor executes a lane keeping assist algorithm.

[0012] According to another aspect of the present disclosure, the state of the vehicle system is determined in response to at least one of vehicle speed, vehicle heading, vehicle curvature, and vehicle direction.

[0013] According to another aspect of the present disclosure, the sensor is a camera for capturing images, and wherein the distance of the host vehicle to the edge of the lane is determined in response to an edge detection algorithm executed on the images.

[0014] According to another aspect of the present disclosure, a method is provided, the method comprising: a vehicle sensor detecting a distance from a host vehicle to an edge of a lane, determining a state of a vehicle system, a processor calculating a risk index in response to the state of the vehicle system, a processor calculating an intervention threshold in response to the risk index, a processor generating a vehicle path in response to a condition that the distance from the host vehicle to the edge of the lane is less than the intervention threshold, and a vehicle controller controlling the vehicle in response to the vehicle path.

[0015] According to another aspect of the present disclosure, the state of the vehicle system is a driver's engagement level determined by a driver monitoring system.

[0016] According to another aspect of the present disclosure, the distance between the intervention threshold and the lane edge increases in proportion to the risk index.

[0017] According to another aspect of the present disclosure, the risk index is determined in response to at least one of pavement material, shoulder type, and road geometry.

[0018] According to another aspect of the present disclosure, the state of the vehicle system is a driver engagement level determined in response to a steering intervention by the driver.

[0019] According to another aspect of the present disclosure, the state of the vehicle system is determined in response to at least one of a vehicle speed and a vehicle direction.

[0020] According to another aspect of the present disclosure, the vehicle sensor is a camera for capturing images, and wherein the distance of the host vehicle to the edge of the lane is determined in response to an edge detection algorithm executed on the images.

[0021] According to another aspect of the present disclosure, the method further includes: the processor executing a lane keeping assist algorithm.

[0022] According to another aspect of the present disclosure, the risk index is calculated in response to a sum of a plurality of weighted risk values ​​determined according to a plurality of system states.

[0023] According to another aspect of the present disclosure, an adaptive driver assistance system for performing an adaptive lane keeping operation in a host vehicle is provided, the adaptive driver assistance system comprising: a camera for capturing a road surface image; an image processor configured to detect lane markings in the image; a processor for estimating a distance between the host vehicle and the lane marking, determining an intervention threshold in response to at least one vehicle system state, and generating a vehicle path in response to a condition that the distance between the host vehicle and the lane marking is less than the intervention threshold; and a vehicle controller for controlling the host vehicle in response to the vehicle path.

[0024] According to another aspect of the present disclosure, the distance between the intervention threshold and the lane edge increases in proportion to the risk index.

[0025] According to another aspect of the present disclosure, the turning radius of the vehicle path increases in proportion to the risk index.

[0026] The above advantages as well as other advantages and features of the present disclosure will become apparent by reading the following detailed description of the preferred embodiments in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other features and advantages of the present invention and the methods for obtaining them will become more obvious and the present invention can be better understood by referring to the following description of the embodiments of the present invention in conjunction with the accompanying drawings, in which:

[0028] Figure 1 An operating environment of an adaptive lane keeping assist system for assisting driving in a motor vehicle is shown for an exemplary embodiment.

[0029] Figure 2 A block diagram of an adaptive lane keeping assist system for assisting driving according to an exemplary embodiment is shown.

[0030] Figure 3 A flow chart of an adaptive lane keeping assist method for assisted driving according to another exemplary embodiment is shown.

[0031] Figure 4 A block diagram of an exemplary embodiment of an adaptive lane keeping assist system for assisting driving is shown.

[0032] Figure 5 A flow chart of an adaptive lane keeping assist method for assisted driving according to another exemplary embodiment is shown.

[0033] The examples given herein are only for illustrating preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention in any way. DETAILED DESCRIPTION

[0034] Several embodiments of the present invention are described herein. However, it should be understood that the disclosed embodiments are exemplary only and that other embodiments may take various alternative forms. The drawings are not necessarily drawn to scale; certain features may be exaggerated or reduced to show details of specific components. Therefore, the specific structural and functional details disclosed herein should not be interpreted as restrictive, but merely representative. The various features shown and described with reference to any one of the figures may be combined with the features shown in one or more other figures to produce embodiments that are not explicitly shown or described. The combination of features shown provides representative embodiments for typical applications. However, various combinations and modifications of features consistent with the teachings of the present disclosure may be desirable for specific applications or implementations.

[0035] Figure 1An operating environment 100 for use of a motor vehicle equipped with an adaptive lane keeping assist system of an exemplary embodiment is schematically illustrated. The exemplary operating environment 100 includes a two-lane road 105 having a central lane divider marking 110 and a gravel shoulder 125. Also illustrated is a host vehicle 115, a host vehicle path 120, and an intervention threshold 122. The host vehicle path 120 is determined in response to a lane keeping assist (LKA) algorithm executed by an advanced driver assistance system (ADAS) within the host vehicle 115. The LKA algorithm monitors lane markings and / or vehicle position, determines an intervention threshold 122, and corrects the vehicle path by adjusting the steering angle to return the vehicle to the center of the current lane of travel when the vehicle approaches the intervention threshold 122, thereby keeping the vehicle within the current lane of travel. Typically, vehicle occupants feel more comfortable using an ADAS system that makes smooth, gradual speed or steering adjustments. The LKA algorithm typically has an intervention threshold 122 close to the edge of the lane of travel, so that when the vehicle reaches the intervention threshold 122, a sudden change in path may be required to prevent the vehicle from exceeding the current lane. As the path angle of the host vehicle approaches the threshold, this abrupt path change may become increasingly uncomfortable, resulting in sharper, more abrupt path changes.

[0036] The exemplary LKA algorithm is configured to implement a new path shaping method for LKA that enables the algorithm to adapt to the driver's preferences as well as the driver's state, vehicle state, and environmental conditions based on a calculated risk index formula. The exemplary ADAS employs an intelligent adaptive formula to calculate a lane departure risk index and utilizes the risk index in automatic path shaping to adaptively keep the vehicle in the lane based on the driver's state, driver behavior, vehicle state, and environmental conditions. Automatic path shaping responsive to the risk index alleviates vehicle occupants' concerns about premature or late LKA intervention. In an exemplary embodiment, the automatic path shaping may adjust the intervention threshold 122 responsive to the calculated risk index.

[0037] Now go to Figure 2 , a block diagram of an exemplary embodiment of an adaptive lane keeping assist system 200 for assisting driving in a motor vehicle is shown. The exemplary system 200 is operable to generate control data to perform an adaptive lane keeping operation on a motor vehicle equipped with ADAS. The exemplary system 200 may include a camera 245, a memory 250 for storing map data, a global positioning system (GPS) 235, an ADAS controller 220, a driver monitoring system (DMS) 240, a vehicle controller 230, a throttle controller 255, a brake controller 260, and a steering controller 270.

[0038] The exemplary system 200 may use one or more cameras 245 as sensors to detect the environment around the host vehicle. For example, the camera 245 may be mounted below each side mirror of the host vehicle to capture images of the road surface on either side of the host vehicle. The ADAS controller 220 or the sensor fusion controller may use image processing techniques such as edge detection to determine the distance from the host vehicle to the lane markings on each side of the vehicle. The exemplary system may also include a memory 250 for storing map data including high-resolution map data, and a GPS 235 for detecting the location of the vehicle. The exemplary system 200 may be configured to estimate the location of the vehicle by combining information received from the GPS 235, the camera 245, and the memory 250.

[0039] The DMS 240 is configured to detect the driver's concentration during vehicle operation. The DMS 240 may include a cabin camera for monitoring the position of the driver's eyes or an infrared sensor for determining the position of the driver's head. The DMS 240 may monitor steering wheel rotation and driving trajectory to determine the driver's engagement in vehicle operation. The DMS 240 may be an active DMS system that requires the driver to interact with the DMS 240 by activating certain vehicle features, responding to DMS prompts, or performing other active actions (e.g., moving the steering wheel, responding to verbal questions, etc.). In this exemplary embodiment, the driver's concentration may be a factor in the risk index formula used to calculate the LKA.

[0040] The ADAS controller 220 first receives data from the one or more cameras 245 and receives data from the GPS 235 and the vehicle controller 230 to estimate the distance from the host vehicle to the edge of the current lane of travel. The edge of the current lane of travel may be determined in response to road lane markings (e.g., white or yellow lines) or in response to a change in material (e.g., from asphalt to gravel). The ADAS controller 220 may use image processing techniques performed on images received from the one or more cameras 245 to estimate the location of the edge of the current lane of travel.

[0041] After determining the distance from the host vehicle to the edge of the road for one or more lateral directions, the ADAS controller 220 can be operated to determine the risk index formula for the LKA algorithm. The ADAS controller 220 can receive data from the GPS 235, the camera 245, the memory 250, and the DMS 240, and use the data to determine the risk index formula. The risk formula index can be determined in response to risk factors, such as environmental conditions (such as road geometry, distance to neighboring objects, and types of neighboring objects) and vehicle conditions (such as speed and steering wheel angle), weather and road conditions, and shoulder types. In addition, driver inputs (such as steering and braking inputs) as well as driver preferences, past driver performance, and choices can also be considered.

[0042] In an exemplary embodiment, to determine the risk formula index, the ADAS processor 220 may combine the driver monitoring system (DMS) and hands on / off steering wheel detection and adjust the intervention according to the driver's state and concentration. By adjusting and providing feedback to the vehicle occupants, the standard mid-term or early intervention information is displayed on the user interface, which can reduce the perception of inconsistent and / or confusing vehicle performance. The ADAS controller 220 can systematically adapt the LKA control signal coupled to the vehicle controller to environmental conditions and driver concentration to improve safety. The exemplary system can provide feedback to the vehicle occupants and reconfigure the LKA for different types of drivers, such as configuring it for a young driver mode. The ADAS controller 220 can combine updated map data to enable detection of built-up areas. The shoulder type can be used to adjust the LKA intervention to prevent the risk of the vehicle hitting the soft shoulder. The ADAS controller 220 can combine road conditions and adjust the LKA intervention according to the risk. The ADAS controller 220 can adjust the risk formula index according to input from the side blind zone and the risk of oncoming vehicles, so that the LKA can intervene earlier on the left lane markings relative to the right lane markings of the road. The ADAS processor 220 and LKA algorithm may learn driver preferences and adjust LKA interventions based on perceived risk to the environment and dynamic factors, such as the desired distance to neighboring vehicles or obstacles.

[0043] The vehicle controller 230 is configured to receive control data from the ADAS controller 220 and control the movement of the host vehicle in response to the control data. The vehicle controller 230 may generate a throttle control signal to couple to the throttle controller 255 to control the speed of the host vehicle. The vehicle controller 230 may generate a brake control signal to couple to the brake controller 260 to control the brake pressure and braking time of the host vehicle. Similarly, the vehicle controller 230 may generate a steering control signal to couple to the steering controller 270 to control the steering direction and path of the host vehicle.

[0044] Now go to Figure 3 , wherein a flow chart of an exemplary embodiment of an adaptive lane keeping assist method 300 for assisted driving in a motor vehicle is shown. During engagement of the LKA algorithm, the method first determines 310 the distance to the edge of the current lane. The edge may be determined for one or both of the left road markings or the right road markings. The road markings may include solid or dashed marking lines or changes in road material, such as from asphalt to gravel, etc. The distance to the road marking may be determined in response to an image captured by a camera mounted on a side of the host vehicle. Image processing techniques such as edge detection may be used to detect road markings in the image and estimate the distance to the road marking, which is then coupled to the ADAS processor.

[0045] The method then receives 315 data from various vehicle systems related to the risk index formula factors. These factors may include driver preferences as well as driver status, vehicle status, and environmental conditions. For example, environmental conditions may include curb type, shoulder type, road friction, construction area road obstacle location, and weather. Additional factors may include driver concentration, road geometry, adjacent object location, lane geometry, vehicle speed, vehicle steering wheel angle, and driver input for inhibiting or overriding braking or steering wheel rotation, etc., determined in response to the DMS.

[0046] The method then determines 320 a risk index for the path shaping algorithm in response to data from the various vehicle systems. Each data received from the vehicle system is assigned a value in response to the risk level associated with each factor. For example, if it is determined that the driver is not paying attention to vehicle operation, a higher value risk factor may be determined. The risk index is then determined in response to the sum of the weighted factors, where each risk value (r i ) has a weight factor (α i ).

[0047]

[0048] The method may then be configured to determine 325 an intervention threshold in response to the risk index. The intervention threshold is a distance from the edge of the lane at which the LKA generates a vehicle path to return the host vehicle to the center of the lane. The distance of the intervention threshold may be determined in response to the risk index such that the intervention threshold is larger in high risk situations and smaller in low risk situations. In an exemplary embodiment, the intervention threshold may be varied in real time in response to the real-time determination of the risk index. Thus, the intervention threshold may be varied continuously during LKA operation.

[0049] The method then compares the distance of the host vehicle to the edge of the lane to an intervention threshold 330. If the distance to the edge of the lane is greater than the intervention threshold, the method returns to re-determine 310 the distance to the edge of the lane. If the distance to the edge of the lane is less than or equal to the intervention threshold, the method is operable to generate a vehicle path for the host vehicle that will be used to control the vehicle back to the center of the lane. In this exemplary embodiment, the vehicle path generated for a high risk index may have a path that returns to the center of the lane more quickly than the vehicle path generated for a low risk index. Alternatively, for a lower risk index, the vehicle path may prevent the vehicle from leaving the lane, but if it is determined that the driver is highly focused on the operation of the host vehicle, the vehicle is not returned to the center of the lane.

[0050] The method then controls 340 the vehicle along the generated vehicle path in response to the generated vehicle path to return the vehicle toward the center of the road lane. The ADAS controller may generate a steering control signal, a throttle control signal, and / or a brake control signal to control the host vehicle along the generated vehicle path. The method then returns to re-determine 310 the distance to the lane edge.

[0051] Now go to Figure 4 , a block diagram of an exemplary embodiment of an adaptive lane keeping assist system 400 for assisting driving in a vehicle is shown. The exemplary system 400 is implemented in a host vehicle with a sensor 410, a processor 420, a vehicle system 430, and a vehicle controller 440.

[0052] In this exemplary embodiment, sensor 410 may be configured to detect the distance from the host vehicle to the edge of the lane. Sensor 410 may be a camera for capturing images of the road surface, including an image processor for performing image detection algorithms (e.g., edge detection, etc.) on the images. Alternatively, sensor 410 may include a LiDAR, radar, infrared transceiver, etc. for determining the distance to objects within the field of view. In an exemplary embodiment, the host vehicle may have multiple cameras with lateral fields of view, so that the lateral distance to the lane markings or the edge of the road can be detected on each side of the host vehicle.

[0053] The exemplary system 400 may also include a vehicle system 430 configured to provide a status of a vehicle system. The vehicle system 430 may include an inertial measurement unit for providing a vehicle acceleration status, a GPS for providing a vehicle position status, and a vehicle sensor for providing a vehicle speed and / or steering direction. The vehicle system 430 may be a sensor for providing a status of a road surface material and / or a shoulder surface material. The vehicle system 430 may be a DMS system for providing a state of concentration of a driver and / or a vehicle occupant or providing a state of a driver type (e.g., a young driver). The vehicle system 430 may include a traction control system for providing a vehicle traction status or for providing an estimate of a road surface friction coefficient. The vehicle system 430 may include a memory for storing maps and map data, and for providing a status of a construction area location, recorded road features and / or faults, and a required lane position offset to avoid blind spots, road fault points, shoulder encroachment points, etc.

[0054] Processor 420 may be an ADAS processor, a digital signal processor, etc., and is configured to determine a risk index in response to a state of vehicle system 430, calculate an intervention threshold in response to the risk index, and generate a vehicle path in response to a condition that the distance from the host vehicle to the edge of the lane is less than the intervention threshold. Initially, processor 420 is configured to execute a lane keeping assist algorithm. In an exemplary embodiment, the risk index is determined in response to at least one of a pavement material, a shoulder type, and a road geometry. The distance between the intervention threshold and the edge of the lane may increase in proportion to the risk index. Similarly, the turning radius of the vehicle path calculated in response to the determination of the vehicle path may decrease in proportion to the risk index.

[0055] The vehicle controller 440 may be used to control the vehicle in response to the vehicle path. The vehicle controller 440 may generate steering control signals, throttle control signals, and brake control signals to couple to the steering controller, throttle controller, and brake controller to control the vehicle along the vehicle path. The vehicle controller 440 may receive data and status updates from other vehicle systems to confirm the position of the host vehicle along the vehicle path.

[0056] In an exemplary embodiment, system 400 may be an adaptive driver assistance system for performing adaptive lane keeping maneuvers in a host vehicle, the system comprising a sensor 410 (e.g., a camera for capturing images of a road surface) and an image processor configured to detect lane markings in the image. The exemplary system may also include a processor 420, such as an ADAS processor, for estimating the distance between the host vehicle and the lane marking, determining an intervention threshold in response to at least one vehicle system state, and generating a vehicle path in response to a condition that the distance between the host vehicle and the lane marking is less than the intervention threshold. Finally, exemplary system 500 may include a vehicle controller 440 for controlling the host vehicle in response to the vehicle path.

[0057] In one exemplary embodiment, the intervention threshold may be determined in response to a previous driver intervention. For example, the system may detect and store previous driver interventions performed by the driver while controlling the vehicle during a previous LKA operation. This may infer the driver's perceived risk to environmental and dynamic factors, such as the expected distance to a neighboring vehicle or obstacle. The system may also include driver feedback provided via a user interface (e.g., an LED indicator light on a dashboard, etc.) in conjunction with the feedback provided to the driver indicating the activation of the intelligent adaptive intervention. The visual indicator light may be configured to provide feedback to the driver of the host vehicle that the system has been activated and / or that the system has adjusted the LKA intervention threshold.

[0058] Now go to Figure 5, wherein a flow chart of an exemplary embodiment of an adaptive lane keeping assist method 500 for assisted driving in a host vehicle is shown. The method first detects 510 the distance from the host vehicle to the edge of the lane. The distance to one or more curbs may be determined in response to GPS and map data, image data, depth maps, etc. In an exemplary embodiment, the distance may be determined in response to an image captured by a camera, and wherein the distance from the host vehicle to the edge of the lane is determined in response to an edge detection algorithm performed on the image.

[0059] The method then determines 520 a state of a vehicle system, such as driver concentration or driver engagement as determined by a driver monitoring system. For example, the state of the vehicle system may be determined in response to at least one of vehicle speed and vehicle direction.

[0060] The method is then configured to cause the processor to calculate 530 a risk index in response to the state of the vehicle system. The risk index may be calculated in response to a plurality of vehicle system states and / or driver input or preference. For example, the risk index may be determined in response to pavement material, shoulder type, and / or road geometry. The risk index may then be calculated in response to a sum of a plurality of weighted risk values ​​determined based on the plurality of system states.

[0061] The method then causes the processor to calculate 540 an intervention threshold in response to the risk index. Typically, the distance between the intervention threshold and the edge of the lane may increase in proportion to the risk index. The method then causes the processor to generate 550 a vehicle path in response to the condition that the distance from the host vehicle to the edge of the lane is less than the intervention threshold. In this example, when the risk index is low, the turning radius of the vehicle path may be reduced in proportion to the risk index, thereby producing a smoother, more comfortable vehicle path. The method then causes the vehicle controller to control 560 the vehicle in response to the vehicle path.

[0062] Although at least one exemplary embodiment is given in the detailed description above, it should be understood that there are many variations. It should also be understood that these exemplary embodiments are merely examples and are not intended to limit the scope, applicability or configuration of the present disclosure in any way. On the contrary, the detailed description above will provide a convenient way for those skilled in the art to implement the one or more exemplary embodiments. It should be understood that various changes may be made to the function and arrangement of the elements without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. An adaptive driving assistance device, comprising: a sensor configured to detect a distance from the host vehicle to an edge of a lane; a vehicle system configured to provide a status of the vehicle system; a processor configured to determine a risk index in response to a state of a vehicle system, calculate an intervention threshold in response to the risk index, and generate a vehicle path in response to the risk index and a condition that a distance of the host vehicle to an edge of a lane is less than the intervention threshold such that a turning radius of the vehicle path decreases in proportion to the risk index, and wherein the state of the vehicle system indicates a level of driver engagement; and; a vehicle controller for controlling the vehicle in response to the vehicle path to execute a lane keeping assist algorithm; The processor is also configured to adjust intervention based on driver status and concentration, provide feedback to vehicle occupants, display standard mid-term or early intervention information on the user interface, and reconfigure lane keeping assist algorithms for different types of drivers.

2. The device according to claim 1, wherein: A driver engagement level determined by a driver monitoring system including a cabin camera for monitoring at least one of a driver's eye position and a driver's head position.

3. The device according to claim 1, wherein: The distance between the intervention threshold and the lane edge increases in proportion to the risk index.

4. The device according to claim 1, wherein: The risk index is determined in response to at least one of pavement material, shoulder type, and road geometry.

5. The device according to claim 1, wherein: The state of the vehicle system is a level of driver engagement determined in response to the driver's steering intervention.

6. The device according to claim 1, wherein: The processor executes a lane keeping assist algorithm.

7. The device of claim 1, wherein: The state of the vehicle system is determined in response to at least one of vehicle speed, vehicle heading, vehicle curvature, and vehicle direction.

8. The device of claim 1, wherein: The sensor is a camera for capturing an image, and wherein the distance of the host vehicle to an edge of a lane is determined in response to an edge detection algorithm executed on the image.

9. A method comprising: The vehicle sensor detects the distance from the host vehicle to the lane edge; determining the status of vehicle systems; a processor calculating a risk index in response to a state of a vehicle system, wherein the state of the vehicle system is indicative of a driver's engagement; The processor calculates an intervention threshold in response to the risk index; The processor generates a vehicle path in response to the risk index and the condition that the distance from the host vehicle to the lane edge is less than the intervention threshold, such that a turning radius of the vehicle path decreases in proportion to the risk index; and The vehicle controller controls the vehicle in response to the vehicle path to execute a lane keeping assist algorithm; The processor is also configured to adjust intervention based on driver status and focus, provide feedback to vehicle occupants, display standard mid-term or early intervention information on the user interface, and reconfigure lane keeping assist algorithms for different types of drivers.

10. The method of claim 9, wherein: A driver engagement level determined by a driver monitoring system including a cabin camera for monitoring at least one of a driver's eye position and a driver's head position.

Images