Improvements to driver assistance systems or related improvements

Abstract

A driver assistance device is provided, configured to: assign an object to a first region of interest (ROI) lane; obtain object data associated with the object traveling in the first ROI lane; and determine, at least based on the object data, that the object is performing a lane change from the first ROI lane to a second ROI lane adjacent to the first ROI lane. A corresponding method is also provided.

Description

[0001] This application is a divisional application of Chinese patent application No. 201880007840.7, filed on February 12, 2018, entitled "Improvements or related improvements to driver assistance systems". Technical Field

[0002] This invention relates to driver assistance systems, and more particularly to means and methods for driver assistance systems for determining lanes. Background Technology

[0003] To avoid accidents and comply with driving regulations, driving a motor vehicle on the road requires drivers to concentrate, often for extended periods. Inattention increases the risk of accidents and / or non-compliance with regulations. Increasingly, driver assistance systems capable of performing auxiliary functions are being installed in driver-controlled vehicles (“autonomous vehicles”). For example, these assistance functions may include reducing some of the driver’s driving responsibilities or monitoring the driver’s performance to anticipate and / or avoid errors.

[0004] Alternatively, driver assistance features may introduce additional capabilities that are not normally available to drivers. For example, such additional capabilities could allow drivers to access more information than they would typically need, enabling them to perform driving tasks more easily. A rear-facing camera that can provide a video feed to the driver while reversing constitutes an example of such an additional capability. In this example, the video feed allows the driver to reverse and park more easily and safely, without necessarily monitoring the driver's performance or performing any tasks for them.

[0005] Therefore, driver assistance systems mitigate the risks to the driver of autonomous vehicles, their passengers, and other road users. Finally, it is believed that driver assistance functions will evolve to the point where they can control most (if not all) aspects of driving an autonomous vehicle. In this scenario, the driver assistance system would be an automated driving system.

[0006] Driver assistance systems may include active devices that can actively intervene in the operation of an autonomous vehicle, such as by changing the vehicle's speed. Driver assistance systems may optionally or additionally include passive devices that, for example, notify the driver of specific driving situations, allowing the user to react to the notification. For example, when an autonomous vehicle unexpectedly deviates from a road marking, the driver assistance system may issue an audible signal. A given autonomous vehicle may include both passive and active systems.

[0007] Typically, a driver assistance system may include at least one sensor. Specific sensors can measure parameters of the vehicle or its surrounding environment. Data from such sensors is processed to draw conclusions based on the sensor measurements. The driver assistance system can then trigger some interaction with the autonomous vehicle or with the driver based on the results of these conclusions.

[0008] Examples of sensors that may be used in driver assistance systems include RADAR systems, LIDAR systems, cameras, vehicle-to-vehicle communication, and vehicle-to-infrastructure communication.

[0009] Driver assistance systems can be used to control various aspects of driving safety or driver monitoring. For example, ACC (“Adaptive Cruise Control”) can use a RADAR or LIDAR system to monitor the distance between the autonomous vehicle and the vehicle directly ahead on the road. Sensors are able to determine the distance to the vehicle ahead. The driver assistance system also knows and can control the speed of the autonomous vehicle. The driver assistance system controls the speed of the autonomous vehicle to maintain a predetermined safe distance relative to the vehicle ahead. For example, the driver assistance system can control the speed to maintain a certain distance between the autonomous vehicle and the vehicle ahead. Alternatively, the driver assistance system can control the speed to maintain a predetermined time interval between the vehicle ahead at a point of passage and the autonomous vehicle passing the same point.

[0010] Existing driver assistance systems monitor the surrounding environment of autonomous vehicles to identify the positions of other vehicles and entities on or around the road in which the autonomous vehicle is traveling. By monitoring the surrounding environment, such driver assistance systems can maintain situational awareness of the autonomous vehicle. This situational awareness can be used to notify the user of potential hazards. For example, when a second vehicle is in the blind spot, or when a second vehicle is detected cutting into the autonomous vehicle's path, the driver can be notified to change lanes. For example, situational awareness can also be used as input to ACC (Adaptive Cruise Control) systems.

[0011] When monitoring the environment surrounding an autonomous vehicle, multiple vehicles and entities can be identified in its vicinity. It is important to select which vehicles or entities (if any) should be used as the basis for the vehicle system's actions or to notify the driver.

[0012] Providing detailed and reliable situational awareness is important for many different driver assistance functions.

[0013] In most driving situations, vehicles travel in designated lanes. That is, roads are divided into many roughly parallel lanes, each forming a path for vehicles to travel on. Sometimes, lanes are designated by road markings on the road surface, which visually indicate the lane boundaries to the driver. Sometimes there are no road markings, and the driver simply needs to be careful not to accidentally enter the lane of oncoming traffic. Sometimes lane markings change along specific road sections. For example, when roads are being worked on, lanes may be narrowed relative to their normal configuration.

[0014] For example, when lane markings are obscured and invisible (e.g., through snow or congested and somewhat chaotic traffic), or when road markings are absent, driver assistance systems cannot identify the position and size of lanes by processing the camera's output. Lane markings may also be obscured by other vehicles or objects, or simply put, the camera cannot visually distinguish lane markings beyond a certain distance from the autonomous vehicle. Summary of the Invention

[0015] One object of the present invention is to provide an improved apparatus for a driver assistance system and a method for operating the apparatus for a driver assistance system, the apparatus attempting to solve some or all of these problems.

[0016] According to a first aspect of the invention, an apparatus for a driver assistance system for a motor vehicle is provided, the apparatus being configured to: generate a list of objects, each object being located near the vehicle, each object having been identified using data from at least one object sensor on the vehicle; search for lane markings on the road on which the vehicle is traveling using data from at least one lane marking sensor on the vehicle; and establish a region of interest (“ROI”) based on at least one detected lane marking, wherein establishing the ROI includes: generating an autonomous lane for the ROI, in which the vehicle is located; generating a left lane for the ROI on the left side of the autonomous lane for the ROI, and generating a right lane for the ROI on the right side of the autonomous lane for the ROI.

[0017] Preferably, the device is also configured to assign at least one object to one of the ROI autonomous lane, the ROI left lane, and the ROI right lane.

[0018] Advantageously, the device is also configured to generate a VIO candidate list, which includes at least one very important object (“VIO”) selected from the assigned objects.

[0019] Conveniently, the VIO candidate list contains the object closest to the vehicle in each of the multiple regions.

[0020] Preferably, the VIO candidate list contains at least one VIO identified as a lane change object.

[0021] Advantageously, each zone is a section of one of the ROI autonomous lane, the ROI right lane, or the ROI left lane.

[0022] Conveniently, the device is configured to send a list of VIO candidates to at least one advanced driver assistance system (“ADAS”).

[0023] Preferably, when exactly one lane marker is identified, the position of the autonomous lane is based on the exactly one lane marker.

[0024] Advantageously, each of the ROI left lane and ROI right lane is a copy of the ROI autonomous lane, which has a lateral offset corresponding to the ROI autonomous lane.

[0025] Conveniently, when an external lane marking is identified, the position of either the left lane or the right lane of the ROI is based on the position of the external lane marking.

[0026] Preferably, the device is also configured to establish ROI based on vehicle parameters even when lane markings are not identified.

[0027] Advantageously, vehicle parameters include at least one of vehicle speed, steering angle, and yaw rate.

[0028] Preferably, the vehicle parameters include the steering angle.

[0029] Advantageously, vehicle parameters include yaw rate and vehicle speed.

[0030] Conveniently, each object includes its position relative to the vehicle.

[0031] According to a second aspect of the invention, a method for a driver assistance system for a motor vehicle is provided, the method comprising the steps of: generating a list of objects, each object being located near a vehicle, each object having been identified using data from at least one object sensor on the vehicle; searching for lane markings on the road on which the vehicle is traveling using data from at least one lane marking sensor on the vehicle; and establishing a region of interest (“ROI”) based on at least one detected lane marking, wherein establishing the ROI comprises: generating an autonomous lane for the ROI, the vehicle being located within the autonomous lane for the ROI; generating a left lane for the ROI on the left side of the autonomous lane for the ROI, and generating a right lane for the ROI on the right side of the autonomous lane for the ROI. Attached Figure Description

[0032] Therefore, the present invention can be more easily understood, and its further features can be understood. Embodiments of the invention will now be described by way of example with reference to the accompanying drawings, in which:

[0033] Figure 1A vehicle having a driver assistance system of the type suitable for the present invention is shown;

[0034] Figure 2 A schematic bird's-eye view of the driving scenario is shown;

[0035] Figure 3 The designated region of interest (“ROI”) according to the present invention is shown;

[0036] Figure 4A A first scenario for generating ROI according to the present invention is shown;

[0037] Figure 4B A second scenario for generating ROI according to the present invention is shown;

[0038] Figure 4C A third scenario for generating ROI according to the present invention is shown;

[0039] Figure 5 The method for generating ROI according to the present invention is illustrated schematically;

[0040] Figure 6 This schematically illustrates a method for assigning objects to an ROI;

[0041] Figure 7 A schematic bird's-eye view of a driving scenario is shown, where objects have been assigned to ROIs;

[0042] Figure 8 A schematic bird's-eye view is shown illustrating a driving scenario of possible lane changes by an autonomous vehicle;

[0043] Figure 9 The method for selecting very important objects (“VIO”) is illustrated schematically;

[0044] Figure 10 A schematic bird's-eye view showing a driving scenario of a vehicle designated as VIO is shown; and

[0045] Figure 11 An overview of the method according to the invention is shown schematically. Detailed Implementation

[0046] Now let's move on to a more detailed consideration. Figure 1 A schematic diagram of an exemplary driver assistance system 1 installed in an autonomous vehicle 2 is shown (where only one side panel is in Figure 1(The text indicates the vehicle's orientation). Safety system 1 includes various types of sensors mounted at appropriate locations on the autonomous vehicle 2. Specifically, the illustrated system 1 includes: a pair of diverging and outward-pointing mid-range radar (“MRR”) sensors 3 mounted at the respective front corners of the vehicle 2; a pair of similar diverging and outward-pointing multi-action radar sensors 4 mounted at the respective rear corners of the vehicle; a forward-pointing long-range radar (“LRR”) sensor 5 mounted at the center of the front of the vehicle 2; and a pair of generally forward-pointing optical sensors 6 forming part of a stereo vision system (“SVS”) 7, which may be mounted, for example, in the upper edge area of ​​the vehicle's windshield. The various sensors 3 to 6 are operatively connected to a central electronic control system, which is typically provided in the form of an integrated electronic control unit 8 mounted in a convenient location within the vehicle. In the specific arrangement shown, the front MRR sensor 3 and the rear MRR sensor 4 are connected to the central control unit 8 via a conventional Controller Area Network (“CAN”) bus 9, and the LRR sensor 5 and the SVS 7 are connected to the central control unit 8 via a faster FlexRay serial bus 9, which is itself a known type.

[0047] Overall, and under the control of control unit 8, various sensors 3 to 6 can be used to provide a variety of different types of driver assistance functions, such as: blind spot monitoring; adaptive cruise control; collision avoidance assist; lane departure protection; and rear collision mitigation. Such systems can be referred to as Advanced Driver Assistance Systems (“ADAS”). These systems constitute part of an automated driving system.

[0048] Figure 2 A bird's-eye view of a typical driving situation on road 15 is shown. Road 15 has three adjacent lanes along which traffic can travel: a central traffic lane 16, a left traffic lane 17, and a right traffic lane 18. The left traffic lane 17 is separated from the central traffic lane 16 by lane marking 19; similarly, the right traffic lane 18 is separated from the central traffic lane 16 by another lane marking 19. The outer boundary of the right traffic lane 17 is specified by a solid lane marking 20; similarly, the outer boundary of the left traffic lane 18 is specified by another solid lane marking 20.

[0049] An autonomous vehicle 21 is shown traveling along the central traffic lane 16. The direction of travel of the autonomous vehicle 21 is indicated by arrow 21A. Three other vehicles 22, 23, and 24 are also shown traveling along road 15. The directions of travel of each of the other vehicles 22, 23, and 24, 22A, 23A, and 24A, are also shown. Figure 1 As shown in the image.

[0050] Obviously, Figure 2All the vehicles shown are traveling in roughly the same direction (upwards). Therefore, for example, Figure 2 Road 15 corresponds to half of a highway or expressway. The lower half of the corresponding highway or expressway is not shown, but will be located on one side of road 15. Figure 2 (As shown on the right or left) (which side obviously depends on the country where the road is located). Vehicles on the lower half of a highway or expressway will generally travel in the opposite direction to vehicles on road 15.

[0051] Figure 3 An autonomous vehicle 21 is shown traveling along the road. Traffic lanes and road markings are shown. Figure 2 The autonomous vehicle 21 includes the device according to the invention.

[0052] As described above, the device uses data from at least one sensor mounted on the autonomous vehicle 21 to search for road markings. The device detected segments of lane markings 19 and 20. That is, in this example, four segments of lane markings 19 and 20 have been detected. These four segments include the segment of each central lane marking 19 and the segment of each outer lane marking 20.

[0053] Based on the detected lane markings, the device is configured to generate a region of interest (“ROI”) near the autonomous vehicle 21. The ROI includes ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27. In this example, ROI autonomous lane 25 is actually the area extending in front of and behind the autonomous vehicle in the central traffic lane (i.e., the lane in which autonomous vehicle 21 is traveling). ROI left lane 26 is actually the area extending in front of and behind the autonomous vehicle in the left traffic lane (i.e., the lane adjacent to the road lane in which autonomous vehicle 21 is traveling). ROI right lane 27 is actually the area extending in front of and behind the autonomous vehicle in the right traffic lane (i.e., the lane in which autonomous vehicle 21 is traveling on the opposite side of the left lane). Figure 3 In the example shown, ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27 correspond to drivable road lanes. However, this is not always the case. In the real world, any of autonomous lane 25, left lane 26, and / or right lane 27 may be infeasible. For example, on a two-lane road with two traffic lanes, there may not be a third drivable lane, but three ROI lanes are still generated. The same applies to single-lane roads—ROIs are always generated. At least one of the ROI lanes may also correspond to a traffic lane on the opposite side of the road leading to the autonomous vehicle, i.e., an oncoming traffic lane.

[0054] Figure 4AThis is a schematic diagram of the first ROI generation scenario. The device has detected four lane markings 19 and 20. ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27 are generated to lie between these four detected lane markings. More specifically, the ROI autonomous lane corresponds to the area defined by the right-hand center lane marking 19 and the left-hand center lane marking 19. ROI left lane 26 corresponds to the area defined by the left-hand center lane marking 19 and the left-hand outer lane marking 20. ROI right lane 27 corresponds to the area defined by the right-hand center lane marking 19 and the right-hand outer lane marking 20. Lane markings 19 and 20 are curved because the road curves to the right in this scenario.

[0055] However, it is not necessary to detect all road markings that define the ROI autonomous lane, ROI left lane, and ROI right lane in order to generate the ROI autonomous lane, ROI left lane, and ROI right lane.

[0056] Figure 4B This is a schematic diagram of the second ROI generation scenario. The device detects two lane markings 19—a left center lane marking and a right center lane marking. The two detected lane markings define the ROI autonomous lane. Therefore, an ROI autonomous lane 25 is generated between the two detected lane markings. The ROI left lane 26 is generated as a copy of the ROI autonomous lane, which has a lateral offset from the ROI autonomous lane (to the left, as shown). Figure 4B (As shown). Accordingly, the ROI right lane 27 is defined as a copy of the ROI autonomous lane, which has a lateral offset from the ROI autonomous lane 25 (lateral offset to the right, as shown). Figure 4B (As shown). The lateral offset of the right lane of the ROI is equal to the lateral offset of the left lane of the ROI. The lateral offset can also be equal to the width of the autonomous lane of the ROI.

[0057] Figure 4C This is a schematic diagram of the third scenario. The device has detected the left center lane marker 19. The single detected left center lane marker defines the left edge of the ROI autonomous lane. The other edge of the ROI autonomous lane is a copy of the detected left center lane marker, which has a lateral offset from the detected lane marker 19. The lateral offset can be the static lane width. The static lane width can be equal to the standard width of the traffic lane. The ROI left lane 26 is defined as a copy of the ROI autonomous lane, which has a lateral offset from the ROI autonomous lane (to the left, as shown in the diagram). Figure 4B (As shown). Accordingly, the ROI right lane 27 is defined as a copy of the ROI autonomous lane, which has a lateral offset from the ROI autonomous lane 25 (lateral offset to the right, as shown). Figure 4B (As shown). The lateral offset of the right lane of the ROI can be equal to the lateral offset of the left lane of the ROI.

[0058] The device can be configured to... Figure 4A , Figure 4B and Figure 4C The operation can be performed in any of the three scenarios mentioned above, depending on how many lane markings are identified.

[0059] If no lane markings are identified, the device can default to the standard operating mode. In the standard operating mode, an ROI autonomous lane is generated, which has a static lane width centered on the autonomous vehicle's position. The curvature of the ROI autonomous lane is determined using autonomous vehicle parameters. These parameters may include at least the steering angle, yaw rate, and vehicle speed. For example, when the speed is below a low threshold, the curvature of the ROI autonomous lane may be defined solely by the steering angle. When the speed is between the low and medium thresholds, the curvature may be defined by both the steering angle and the yaw rate. When the speed is above a medium threshold, the curvature may be defined by both the yaw rate and the speed. For example, the low threshold may be 12 km / h. For example, the medium threshold may be 30 km / h. As described above, the left and right autonomous lanes can be generated as copies of the ROI autonomous lane with corresponding lateral offsets.

[0060] Figure 5 A method implemented by the apparatus according to the invention is illustrated. The method includes input data 30 processed by an ROI builder 31 to generate ROI lanes 32.

[0061] Input data 30 may include lane information, which describes all detected lane markings. Lane marking information can be provided in various formats. Importantly, the lane information describes the location of the corresponding lane markings in the real world. For example, lane information can be the output of a lane fusion module.

[0062] Input data 30 may also include autonomous vehicle data. Autonomous vehicle data may include parameters describing the current position of the autonomous vehicle and its control. For example, vehicle yaw rate, vehicle speed, and vehicle steering angle may be included in the autonomous vehicle data.

[0063] The input data may also include driver intent data. Driver intent data includes information that can be used to determine the actions the driver might take before actually performing them. For example, driver intent data may include whether the driver is using an indicator that would indicate an upcoming turn or lane change.

[0064] Figure 6 A method that can be implemented by the apparatus according to the invention is shown. The process includes ROI allocation input data 33, which is processed by an object-to-ROI allocation process 34 to generate ROI-allocated objects 35.

[0065] Input data 33 includes data from... Figure 5 The ROI lanes generated by the ROI builder shown are illustrated. Input data 33 also includes object data for each of several objects near the autonomous vehicle. Object path data may include object history, which includes a historical record of the paths taken by the respective object. Object data may also include object path predictions for the respective object, which include predictions of the paths the respective object may take in the future. Object path predictions may include the probability of the object changing lanes (e.g., from the right lane or left lane of the ROI to the autonomous lane of the ROI). Object data may also include the current location of the object. Objects are assigned to ROIs based on the object data and the ROI lanes. Specifically, each object is assigned to a specific ROI lane. For example, a particular object may be assigned to one of the autonomous lane, the right lane, or the left lane of the ROI.

[0066] Objects located outside the ROI can be discarded, ignored, or not processed further in the device. The device can be configured to recognize objects located outside the ROI, but then ignore them.

[0067] Figure 7 It shows that according to Figure 6 The steps shown are a schematic bird's-eye view of the driving scenario after assigning multiple objects to the ROI. An autonomous vehicle 21 is shown traveling along the central traffic lane of the road. As described above, ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27 are generated. Three other vehicles are shown near autonomous vehicle 21.

[0068] • The first vehicle, 36, has been assigned to the left lane of the ROI, 26.

[0069] • The second vehicle, 37, has been assigned to the right lane of the ROI, 27.

[0070] • A third vehicle 38, which has been designated as a vehicle changing lanes, may be assigned information that includes the ROI lane in which the third vehicle is changing lanes. For example, the third vehicle is changing from the right lane of an ROI to an autonomous lane of an ROI.

[0071] The third vehicle 38 can be initially assigned a lane to the ROI, but that lane can change as the third vehicle changes lanes.

[0072] In this example, the initially assigned ROI lane could be ROI right lane 27.

[0073] Determining whether a vehicle is changing lanes can be based on many factors, including:

[0074] • Object path prediction; in other words, does object path prediction include transitions between ROI lanes?

[0075] • Lateral acceleration and / or lateral velocity of the object; in other words, the velocity or acceleration component of the object in the direction of the adjacent ROI lane from the lane where the object is located, which can be used to assess whether the object is changing lanes.

[0076] • The distance between the object and the line extending along the center of the ROI autonomous lane. For example, if a vehicle is close to the center line but is assigned to the right lane of the ROI, it can indicate that the vehicle is making a lane change from the right lane of the ROI to the autonomous lane of the ROI.

[0077] Vehicles changing lanes between two ROI lanes can also be identified by determining the time interval after a vehicle crosses a road marker between the two ROI lanes. When this time interval exceeds a predetermined lane-changing threshold, it can be determined that the vehicle is changing lanes or has already changed lanes. The predetermined lane-changing threshold can be between 1 and 4 seconds. For example, the predetermined lane-changing threshold could be equal to 1.5 seconds.

[0078] The allocation of each vehicle near autonomous vehicle 21 can be based on many factors, such as:

[0079] • Lane shape (e.g., curvature);

[0080] Lane width;

[0081] • The location of objects within the ROI;

[0082] • Object location history;

[0083] • Object path prediction;

[0084] Figure 8 A schematic bird's-eye view of a possible lane-changing scenario is shown. This diagram illustrates how the device is configured to determine and react to lane changes by the autonomous vehicle 21. The autonomous vehicle 21 initially travels along the central traffic lane. As described above, ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27 are generated. Autonomous vehicle data includes vehicle yaw rate and steering angle, as well as the corresponding positions of ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27. As the autonomous vehicle 21 begins to move away from the center of ROI autonomous lane 25 (e.g., ...), the lane changes occur. Figure 6As shown, there are two possible outcomes. The aborted lane change, shown by abort path 39, is the first option. The completed lane change, shown by complete path 40, is the second option. The device can be configured to detect lane changes. A lane change from an ROI autonomous lane to an ROI right lane (for example) causes the previous ROI right lane to be reassigned as an ROI autonomous lane. The previous ROI autonomous lane becomes an ROI left lane, and a new ROI right lane is generated. Of course, for an autonomous vehicle's transition from an ROI autonomous lane to an ROI left lane, the corresponding ROI lane generation and reassignment operations are also possible.

[0085] Driver intent data can be used to determine the likelihood of autonomous vehicle 21 moving away from the center of the ROI autonomous lane, thereby causing autonomous vehicle 21 to complete a lane change.

[0086] Figure 9 A method that can be implemented by the apparatus according to the invention is shown. The process includes VIO selection input data 41, which is processed by VIO selection process 42 to form VIO selection output data 43.

[0087] VIO selection input data 41 may include objects already assigned to ROI lanes and ROI lanes. The VIO selection process selects a subset of the assigned objects as VIO candidates, and may also select different subsets of the objects as fixed VIOs to the fixed VIO list.

[0088] A single VIO candidate can be selected from each ROI lane. The selected VIO candidate can be the object closest to the autonomous vehicle. Alternatively, multiple objects from each ROI lane can be selected as VIO candidates.

[0089] You can assign a VIO candidate type to each VIO candidate. You can specify seven VIO candidate types, for example:

[0090] • Ahead, this means the closest object assigned to the ROI autonomous lane and located in front of the autonomous vehicle;

[0091] • Head, which refers to the object assigned to the autonomous lane of the ROI and located directly in front of the "forward" object;

[0092] • Behind, this means the closest object assigned to the ROI autonomous lane and located behind the autonomous vehicle;

[0093] • Left front, which means the closest object assigned to the left lane of the ROI and located in front of the autonomous vehicle;

[0094] • Right front, which means the closest object assigned to the right lane of the ROI and located in front of the autonomous vehicle;

[0095] • Left rear, which means the closest object assigned to the left lane of the ROI and located behind the autonomous vehicle;

[0096] • Right rear, which means the closest object assigned to the right lane of the ROI and located behind the autonomous vehicle.

[0097] Each object input to the VIO selection process can include a motion type parameter. Several possible values ​​for the motion type parameter can be defined. For example:

[0098] • Unknown, including the type of movement;

[0099] • Driving, in which the object moves in the same direction as the autonomous vehicle;

[0100] • Approaching head-on, with the object moving towards the autonomous vehicle;

[0101] • Stopped, where the object is stationary but has previously been observed moving; and

[0102] • Reverse, where the object was previously seen moving in the direction of the autonomous vehicle, but is now moving in the opposite direction.

[0103] Objects possessing one of these sports types can be entered into the VIP candidate list.

[0104] Objects can also have a "stationary" movement type. Such objects can be entered into a fixed list of VIOs.

[0105] VIO selection input data 41 may also include the vehicle type for each object. Exemplary vehicle types may include “Unknown,” “Car,” “Truck,” and “Trailer.” Any of these vehicle types can be entered into a fixed VIO list or a VIO candidate list.

[0106] There may be distance limitations beyond which it becomes practically impossible to determine the vehicle type of an object. For objects beyond the distance limit, only the object's motion type can be monitored and used in VIO selection. The distance limit can be determined by the capabilities of the sensors on the autonomous vehicle. For example, the distance limit could be 50 meters.

[0107] Figure 10 A schematic bird's-eye view of another driving scenario is shown. This scenario illustrates the selection of very important objects (“VIOs”) from those assigned to ROIs.

[0108] exist Figure 10In this diagram, autonomous vehicle 21 travels along the central traffic lane of the road segment. As described above, ROI autonomous lane 25, ROI left lane 26, and ROI right lane 27 are generated. Six other vehicles traveling on the road segment are also shown. These six vehicles are as follows:

[0109] • The first VIO vehicle 44 is the selected VIO candidate after the left of the VIO type;

[0110] • The second VIO vehicle 45 is the selected VIO candidate to the left of the VIO type;

[0111] • The third VIO vehicle 46 is the selected VIO candidate in the VIO type head;

[0112] • The fourth VIO vehicle 47 is the selected VIO candidate preceding the VIO type;

[0113] • The fifth VIO vehicle 48 is the selected VIO candidate to the right of the VIO type;

[0114] • The sixth VIO vehicle 49 is the selected VIO candidate to the right of the VIO type.

[0115] The first to sixth vehicles are in the VIO candidate list. The VIO candidate list and / or fixed VIO list can be used as input to downstream driver assistance / autonomous driving systems. Each downstream system can use a VIO according to the specific function of the system. Therefore, the device according to the invention performs VIO identification only once. Vehicles equipped with such a device can be controlled in various ways based on the ROI according to the invention. The device can be configured to send VIOs (including the VIO candidate list and / or fixed VIO list) to at least one downstream system. The device can be configured to send VIOs (including the VIO candidate list and / or fixed VIO list) to more than one downstream system. The device according to the invention can be configured to make VIOs (including the VIO candidate list and / or fixed VIO list) accessible and usable by downstream systems. The device can be configured to calculate the distance between each object and the autonomous vehicle along the ROI lane to which the object is assigned. These distances can be used to select a VIO.

[0116] Figure 11An overview of a method that can be implemented by the apparatus according to the invention is shown. The method begins by generating ROIs using an ROI builder 50 according to the method described above. The output 51 of the ROI builder includes ROIs (i.e., an autonomous ROI lane, a left ROI lane, and a right ROI lane). ROI lanes are input into objects in an ROI allocation stage 52, where detected objects located near the autonomous vehicle are assigned to ROIs. The output 53 of the ROI allocation stage 52 includes ROI-assigned objects (i.e., objects already assigned to ROI lanes). The ROI-assigned objects are input into a VIO selection stage 54, where a VIO candidate list and a fixed VIO list are generated. These lists can then be used in autonomous driving and / or driver assistance systems to control the vehicle and / or generate interactions with the driver of the autonomous vehicle.

[0117] When used in this specification and claims, the terms "comprising" and "including," and variations thereof, mean to include the specified features, steps, or integers. These terms should not be construed as excluding the presence of other features, steps, or integers.

[0118] The features disclosed in the foregoing description or the subsequent claims or drawings, expressed in their particular form or in the manner of means for performing the disclosed functions or methods or processes for obtaining the disclosed results, may, as appropriate, be used alone or in any combination of these features to implement the invention in its various forms.

[0119] While the invention has been described in conjunction with the foregoing exemplary embodiments, many equivalent modifications and variations will be apparent to those skilled in the art upon presentation of this disclosure. Therefore, the exemplary embodiments of the invention set forth above are to be considered illustrative rather than restrictive. Various changes may be made to the described embodiments without departing from the spirit and scope of the invention.

Claims

1. A driver assistance device for a vehicle, the device being configured to: Objects detected near the vehicle are assigned to a first region of interest lane, wherein... The first region of interest lane is adjacent to the second region of interest lane, and the vehicle is traveling in the second region of interest lane; Obtain object data associated with the object's movement in the lane of the first region of interest; Based on the object data, determine path prediction information for the object; as well as Based at least on the path prediction information, it is determined that the object is undergoing a lane change from the first region of interest lane to the second region of interest lane.

2. The apparatus according to claim 1, wherein, The path prediction information indicates the transition between lanes in the first region of interest and lanes in the second region of interest.

3. The apparatus according to claim 1, wherein, The device is also configured to: Based on the object data, determine at least one of the lateral acceleration or lateral velocity of the object in the direction of the second region of interest lane from the first region of interest lane; as well as The object is determined to be changing lanes from the first region of interest lane to the second region of interest lane based at least on at least one of the lateral acceleration or the lateral velocity.

4. The apparatus according to claim 1, wherein, The device is also configured to: The distance between the object and the centerline of the lane along the second region of interest is determined based on the object data; as well as The distance indicates that the object is approaching the centerline along the second region of interest lane, thus determining that the object is making a lane change from the first region of interest lane to the second region of interest lane.

5. The apparatus according to claim 1, wherein, The device is also configured to: The time period following the object crossing the road markings between the first region of interest lane and the second region of interest lane is determined based on the object data; as well as The object is determined to be changing lanes from the first region of interest lane to the second region of interest lane, at least based on the time period exceeding a predetermined lane change threshold.

6. The apparatus according to claim 1, wherein, The device is also configured to: The distance between the object and the centerline of the lane along the first region of interest is determined based on the object data; as well as The distance indicates that the object is moving away from the centerline along the first region of interest lane, thus determining that the object is changing lanes from the first region of interest lane to the second region of interest lane.

7. The apparatus according to claim 1, wherein, The device is also configured to: Based on the object data, driver intention data for the object is determined, which indicates the likelihood that the object will move away from the centerline of the lane along the first region of interest. as well as The determination is based at least on the driver intent data that the object is changing lanes from the first region of interest lane to the second region of interest lane.

8. A method for driver assistance in a vehicle, comprising: Objects detected near the vehicle are assigned to a first region of interest lane, wherein the first region of interest lane is adjacent to a second region of interest lane, and the vehicle is traveling in the second region of interest lane; Obtain object data associated with the object's movement in the lane of the first region of interest; Based on the object data, determine path prediction information for the object; and Based at least on the path prediction information, it is determined that the object is undergoing a lane change from the first region of interest lane to the second region of interest lane.

9. The method according to claim 8, wherein, The path prediction information indicates the transition between lanes in the first region of interest and lanes in the second region of interest.

10. The method of claim 8, further comprising: Based on the object data, determine at least one of the lateral acceleration or lateral velocity of the object in the direction of the second region of interest lane from the first region of interest lane; as well as The object is determined to be changing lanes from the first region of interest lane to the second region of interest lane based at least on at least one of the lateral acceleration or the lateral velocity.

11. The method of claim 8, further comprising: The distance between the object and the centerline of the lane along the second region of interest is determined based on the object data; as well as The distance indicates that the object is approaching the centerline along the second region of interest lane, thus determining that the object is making a lane change from the first region of interest lane to the second region of interest lane.

12. The method according to claim 8, further comprising: The time period following the object crossing the road markings between the first region of interest lane and the second region of interest lane is determined based on the object data; as well as The object is determined to be changing lanes from the first region of interest lane to the second region of interest lane, at least based on the time period exceeding a predetermined lane change threshold.

13. The method of claim 8, further comprising: The distance between the object and the centerline of the lane along the first region of interest is determined based on the object data; as well as The distance indicates that the object is moving away from the centerline along the first region of interest lane, thus determining that the object is changing lanes from the first region of interest lane to the second region of interest lane.

14. The method of claim 8, further comprising: Based on the object data, driver intention data for the object is determined, which indicates the likelihood that the object will move away from the centerline of the lane along the first region of interest. as well as The determination is based at least on the driver intent data that the object is changing lanes from the first region of interest lane to the second region of interest lane.

Images