Process-implemented method in a vehicle for detecting and tracking objects using radar data

Abstract

Processor-implemented method in a vehicle (100) for detecting and tracking objects using radar data, the method comprising: Retrieval of radar measurements taken at various periodic time steps by a radar system in the vehicle (100); Organizing the radar measurements as time-ordered groups of radar measurements by the processor (44) into suitable time windows, wherein the time window in which a time-ordered cluster of radar measurements is organized corresponds to the time span in which the time-ordered cluster of radar measurements was carried out; Building a sequence cluster of radar measurements by the processor (44), wherein the sequence cluster comprises several time-ordered clusters of radar measurements corresponding to a first object and the several time-ordered clusters in the sequence cluster are arranged in chronological order, wherein the sequence cluster is arranged as a sliding window of radar measurements and the sliding window includes a predetermined number of the most recent time windows of radar measurements; Removal of noise from the sequence cluster of radar measurements by the processor (44) by removing a cluster of radar measurements from the sequence cluster of radar measurements that is inconsistent with a street topology map for an area where the first object is estimated to be localized; and Output the sequence cluster of radar measurements after removing contradictory radar measurements as a new cluster of radar measurements; removing noise from the sequence cluster of radar measurements includes the following: Comparing the sequence cluster of radar measurements with the road topology map, wherein the road topology map is configured to identify lanes and permissible directions of travel within the identified lanes; and Removal of radar measurements indicating object movements in a direction that contradicts a permissible direction of travel in the lanes of the road topology map.

Description

TECHNICAL AREA

[0001] The present invention relates generally to object detection and tracking, and in particular to systems and methods in a vehicle for using lane information to limit radar tracks used in object detection and tracking. More specifically, the invention relates to a processor-implemented method in a vehicle for detecting and tracking objects using radar data.

[0002] For general background information, reference should be made here to EP 3 401 182 A1, which, however, was not yet known at the relevant time priority for the present invention. BACKGROUND

[0003] Vehicle perception systems have been integrated into vehicles to enable them to perceive their surroundings and, in some cases, to navigate autonomously or semi-autonomously. Sensors that can be used in vehicle perception systems include radar, lidar, image sensors, and others.

[0004] While significant progress has been made in vehicle perception systems in recent years, such systems could still be improved in several ways. For example, radar measurements contain noise from static object reflections, such as the ground, power lines, manhole covers, and the like, as well as atmospheric noise. Moving objects detected by radar measurements may also not appear to be in a single location. They may appear to be spread out over a larger area. Furthermore, background noise can cause positional shifts in objects detected by radar measurements. Consequently, relying solely on motion reflections may not be effective.

[0005] Accordingly, it is desirable to have improved systems and methods for determining returns corresponding to moving objects. Furthermore, other desirable functions and features of the present invention will become apparent from the following detailed description and the attached claims, in conjunction with the attached drawings, as well as with the preceding technical field and background. SUMMARY

[0006] According to the invention, a processor-implemented method in a vehicle for detecting and tracking objects using radar data is proposed, wherein the method is characterized by the features of claim 1.

[0007] Systems and methods for an improved object detection and tracking system in a vehicle are provided. In one embodiment, a processor-implemented method in a vehicle for detecting and tracking objects using radar data includes the retrieval by the processor of radar measurements taken at various periodic time steps by a radar system in the vehicle, wherein the radar measurements are organized by the processor as time-ordered groups of radar measurements in suitable time windows, the time window in which a time-ordered cluster of radar measurements is organized corresponding to the time interval in which the time-ordered cluster of radar measurements was carried out.The process involves the processor constructing a sequence cluster of radar measurements, wherein the sequence cluster comprises multiple time-ordered clusters of radar measurements corresponding to a first object, and the multiple time-ordered clusters in the sequence cluster are arranged in chronological order, the sequence cluster being arranged as a sliding window of radar measurements, and the sliding window containing a predetermined number of the most recent time windows of radar measurements. The method further comprises the processor removing noise from the sequence cluster of radar measurements by removing a cluster of radar measurements from the sequence cluster that contradicts a road topology map for an area in which the first object located there is estimated; and outputting the sequence cluster of radar measurements after the removal of contradictory radar measurements as a new cluster of radar measurements.

[0008] According to the invention, removing noise from the sequence cluster of radar measurements involves comparing the sequence cluster of radar measurements with the road topology map, wherein the road topology map is configured to identify lanes and permissible directions of travel in the identified lanes; and removing radar measurements that indicate object movements in a direction that contradicts a permissible direction of travel in the lanes of the road topology map.

[0009] In one embodiment, the method further comprises: tracking one or more radar tracks using a separate instance of a restricted Kalman filter for each radar track, wherein each radar track comprises successive observations of the same object, and wherein the tracking involves forcing and restricting the movement of the one or more radar tracks within an aligned track in a manner dictated by the aligned track, even if a measurement supporting movement is missing in the one or more radar tracks; determining whether the new cluster of radar measurements can correspond to a detected object aligned within a lane by comparing the new cluster of radar measurements with the radar tracks aligned within a lane;and not assigning the new cluster of radar measurements to one of the radar tracks if the comparison does not lead to the identification of a radar track to which the new cluster can correspond.

[0010] In one embodiment, the method further includes: searching in aligned tracks for previous radar measurements corresponding to the identified radar track that match the new cluster of radar measurements, if the comparison leads to the identification of a radar track to which the new cluster may correspond; and assigning the new cluster of radar measurements to the identified radar track if the previous radar measurements corresponding to the identified radar tracks match the new cluster of radar measurements.

[0011] In one embodiment, the method further includes rejecting the new cluster of radar measurements as noise if the earlier radar measurements corresponding to the identified radar tracks do not match the new cluster of radar measurements.

[0012] In a further embodiment, a processor-implemented method in a vehicle for detecting and tracking objects using radar data comprises: tracking one or more radar tracks using a separate instance of a restricted filter for each radar track, wherein each radar track includes successive observations of the same object, and wherein the tracking involves forcing and restricting the movement of the one or more radar tracks within an aligned track in a manner dictated by the aligned track, even if a measurement supporting movement is missing in the one or more radar tracks; determining whether a new cluster of radar measurements can correspond to a detected object aligned within a lane by comparing the new cluster of radar measurements with the radar tracks aligned within a lane;Searching aligned tracks for previous radar measurements corresponding to the identified radar track that match the new cluster of radar measurements, if the comparison leads to the identification of a radar track that the new cluster can match; and assigning the new cluster of radar measurements to the identified radar track if the previous radar measurements corresponding to the identified radar track match the new cluster of radar measurements.

[0013] In one embodiment, the method further includes predicting a future observation for the identified radar track by projecting the future observation onto a path specified by the aligned track.

[0014] In one embodiment, the prediction of a future observation occurs when an obscuring object prevents the vehicle from receiving radar feedback from sections of the aligned lane.

[0015] In one embodiment, the prediction of a future observation includes the prediction that the detected object may cross a different track.

[0016] In one embodiment, the prediction of a future observation includes the prediction that the detected object may move into a different track due to the requirements of the aligned track.

[0017] In one embodiment, the aligned lane includes a pure curved section of the roadway.

[0018] In one embodiment, the aligned lane includes a pure lane section.

[0019] In one embodiment, the aligned lane includes a section of roadway without the need for a U-turn.

[0020] In one embodiment, the method further includes generating the new cluster of radar measurements by removing a cluster of radar measurements that contradicts a road topology map.

[0021] In one embodiment, removing a cluster of radar measurements that contradicts a road topology map involves comparing the sequence cluster of radar measurements with the road topology map, wherein the road topology map is configured to identify lanes and permissible directions of travel in the identified lanes; and removing radar measurements that indicate object movement in a direction that contradicts a permissible direction of travel in the lanes in the road topology map.

[0022] In another embodiment, a vehicle includes: a radar system configured to generate radar data; and an object detection system comprising one or more processors configured by programming instructions in non-volatile, computer-readable media.The object detection system is configured to: retrieve radar measurements taken at various periodic time steps by a radar system in the vehicle; organize the radar measurements into suitable time windows, wherein the time window into which a set of radar measurements is organized corresponds to the time span in which the set of radar measurements was taken; construct a sequence cluster of radar measurements, wherein the sequence cluster includes radar measurements corresponding to a first object in a multitude of different time windows, wherein the sequence cluster includes a sliding window of radar measurements, and the sliding window includes a predetermined number of the most recent time windows of radar measurements.The object detection system is further configured to remove noise from the sequence cluster of radar measurements by removing a cluster of radar measurements from the sequence cluster of radar measurements that contradicts a road topology map for an area where the first object is estimated to be located; and outputting the sequence cluster of radar measurements after removing contradictory radar measurements as a new cluster of radar measurements.

[0023] In one embodiment, the object detection system is further configured to: compare the sequence cluster of radar measurements with the road topology map, wherein the road topology map is configured to identify lanes and permissible directions of travel in the identified lanes; and remove radar measurements that indicate object movements in a direction that contradicts a permissible direction of travel in the lanes of the road topology map.

[0024] In one embodiment, the vehicle further includes an object tracking system. The object tracking system includes one or more processors configured by programming instructions in non-volatile, computer-readable media. The object tracking system is configured to: track one or more radar tracks using a separate instance of a restricted filter for each radar track, wherein each radar track includes successive observations of the same object, and wherein the tracking involves forcing and restricting the movement of the one or more radar tracks within an aligned track in a manner dictated by the aligned track, even if a measurement supporting movement is absent in the one or more radar tracks;Determine whether a new cluster of radar measurements can correspond to a detected object aligned within a lane by comparing the new cluster of radar measurements with the radar tracks aligned within a lane; search in aligned tracks for previous radar measurements corresponding to the identified radar track that correspond to the new cluster of radar measurements if the comparison leads to the identification of a radar track that the new cluster can correspond to; and assign the new cluster of radar measurements to the identified radar track if the previous radar measurements corresponding to the identified radar track correspond to the new cluster of radar measurements.

[0025] In one embodiment, the object tracking system is further configured to predict a future observation for the identified radar track by projecting the future observation onto a path specified by the aligned track.

[0026] In one embodiment, the object tracking system is further configured to reject the new cluster of radar measurements as noise if the earlier radar measurements corresponding to the identified radar tracks do not match the new cluster of radar measurements. DESCRIPTION OF THE DRAWINGS

[0027] The exemplary embodiments are described below in conjunction with the following drawings, where the same reference numerals denote the same elements, and where the following applies: Fig. Figure 1 shows an exemplary vehicle incorporating an object detection system and an object tracking system for use with a radar system according to various embodiments; Fig. Figure 2 is a functional block diagram illustrating an autonomous driving system (ADS) in conjunction with an autonomous vehicle according to various embodiments; Fig. Figure 3 is a block diagram of an exemplary object detection and tracking system in an exemplary vehicle according to various embodiments; Fig. Figure 4A is a diagram illustrating exemplary radar measurements in the time intervals t0, t1 and t2 according to different embodiments; Fig. Figure 4B is a diagram illustrating the assignment of the exemplary radar measurements in the time intervals t0, t1 and t2 to an exemplary measurement cluster and a direction of movement for the cluster according to different embodiments; Fig. 4C is a diagram that shows an exemplary superimposition of lanes on the cluster according to different embodiments; Fig. 4D is a diagram that depicts a further exemplary superimposition of lanes on the cluster according to different embodiments; Fig. Figure 5A is a diagram illustrating exemplary radar measurements at time intervals t0, t1, t2, t3, t4 and t5 according to different embodiments; Fig. 5B is a diagram illustrating the direction of travel of the exemplary radar measurements in the time intervals t0, t1, t2, t3, t4 and t5 according to different embodiments; Fig. 5C is a diagram that shows an exemplary superimposition of lanes onto the exemplary radar measurements according to different embodiments; Fig. Figure 6A is a diagram showing an exemplary cluster of radar measurements taken at time steps t0, t1, t2 and t3 with respect to the exemplary oriented lanes according to different embodiments; Fig. Figure 6B is a diagram showing an exemplary cluster of radar measurements taken at time steps t0, t1 and t2 with respect to exemplary oriented lanes according to different embodiments; Fig. Figure 7A is a diagram illustrating exemplary radar measurements in time steps t0, t1 and t2 with respect to an object in an example track according to different embodiments; Fig. Figure 7B is a diagram illustrating exemplary radar measurements in time steps t0, t1 and t2 with respect to an object in an exemplary track according to different embodiments; Fig. Figure 7C is a diagram illustrating exemplary radar measurements in time steps t0, t1 and t2 with respect to an object in an exemplary track according to various embodiments; and Fig. Figure 8 is a process flow diagram that illustrates an exemplary process for object detection and tracking using radar measurements according to various embodiments. DETAILED DESCRIPTION

[0028] The following detailed description serves only as an example. Furthermore, there is no intention to be bound by any theory explicitly or implicitly presented in the preceding technical section, background, summary, or the following detailed description. The term "module" as used herein refers to all hardware, software, firmware products, electronic control components, processing logic, and / or processor devices, individually or in any combination, including, but not limited to, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), an electronic circuit, a processor (shared, dedicated, or group processor), and memory executing one or more software or firmware programs, a combinational logic circuit, and / or other suitable components providing the described functionality.

[0029] Embodiments of the present invention may be described herein as functional and / or logical block components and various processing steps. It should be noted that such block components may be composed of any number of hardware, software, and / or firmware components configured to perform the required functions. For example, an embodiment of the present invention of a system or component may employ various integrated circuit components, such as memory elements, digital signal processing elements, logic elements, value tables, or the like, which can perform multiple functions under the control of one or more microprocessors or other control devices.Furthermore, experts in the field will recognize that the exemplary embodiments of the present invention can be used in conjunction with any number of systems, and that the system described herein is merely an exemplary embodiment of the present invention.

[0030] For the sake of brevity, conventional techniques related to signal processing, data transmission, signal generation, control, machine learning models, radar, lidar, image analysis, and other functional aspects of the systems (and the individual operating components of the systems) cannot be described in detail herein. Furthermore, the connecting lines shown in the various figures are intended to represent exemplary functional relationships and / or physical connections between the various elements. It should be noted that many alternative or additional functional relationships or physical connections may exist in an embodiment of the present invention.

[0031] Fig. Figure 1 shows an exemplary vehicle 100, which includes an object detection and an object tracking system 302 for use with a radar system. As in Fig. As shown in Figure 1, the vehicle 100 generally comprises a chassis 12, a body 14, front wheels 16, and rear wheels 18. The body 14 is mounted on the chassis 12 and essentially encloses the other components of the vehicle 100. The body 14 and the chassis 12 can together form a frame. The wheels 16-18 are each rotatably coupled to the chassis 12 near a respective corner of the body 14.

[0032] In various embodiments, the vehicle 100 can be an autonomous or semi-autonomous vehicle. For example, an autonomous vehicle 100 is a vehicle that is automatically controlled to transport passengers from one place to another. In the illustrated embodiment, the vehicle 100 is depicted as a passenger car, but any other vehicle, including motorcycles, trucks, sports vehicles (SUVs), recreational vehicles (RVs), ships, aircraft, etc., can also be used.

[0033] In an exemplary embodiment, the vehicle 100 can correspond to a Level Four or Level Five automation system according to the Society of Automotive Engineers (SAE) Standard Taxonomy of Automated Driving Levels “J3016”. Using this terminology, a Level Four system denotes a “high degree of automation” with reference to a driving mode in which the automated driving system performs all aspects of the dynamic driving task, even if a human driver does not respond appropriately to a request for intervention. A Level Five system, on the other hand, exhibits “full automation” and denotes a driving mode in which the automated driving system performs all aspects of the dynamic driving task under all road and environmental conditions that a human driver can handle.It is understood, however, that the embodiments described in this document are not limited to a specific taxonomy or category of automation. Furthermore, systems according to this embodiment can be used in conjunction with any vehicle in which the present subject matter can be implemented, regardless of its level of autonomy.

[0034] As shown, the vehicle 100 generally includes a drive system 20, a transmission system 22, a steering system 24, a braking system 26, a sensor system 28, an actuator system 30, at least one data storage device 32, at least one control unit 34, and a communication system 36. The drive system 20 may, in various embodiments, include an internal combustion engine, an electric machine such as a traction motor, and / or a fuel cell drive system. The transmission system 22 is configured to transmit power from the drive system 20 to the vehicle wheels 16 and 18 according to the selectable gear ratios. According to various embodiments, the transmission system 22 may include a gear-ratio automatic transmission, a continuously variable transmission, or another suitable transmission.

[0035] The braking system 26 is configured to provide a braking torque to the vehicle wheels 16 and 18. The braking system 26 can include, in various embodiments, friction brakes, bake-by-wire, a regenerative braking system such as an electric motor, and / or other suitable braking systems.

[0036] The steering system 24 influences the position of the vehicle wheels 16 and / or 18. While in some embodiments a steering wheel 25 is shown, the steering system 24 may not include a steering wheel.

[0037] The sensor system 28 includes one or more sensor devices 40a-40n that detect observable conditions of the external environment and / or the internal environment of the vehicle 100 (for example, the condition of one or more occupants) and generate corresponding sensor data. Sensor devices 40a-40n may include, but are not limited to, radars (e.g., long-range, medium-range, short-range), lidar, global positioning systems, optical cameras (e.g., forward-facing, 360-degree, rear-facing, side-facing, stereo, etc.), thermal imaging cameras (e.g., infrared), ultrasonic sensors, velocity sensors (e.g., encoders), and / or other sensors that can be used in conjunction with systems and methods according to the present subject matter.

[0038] The actuator system 30 includes one or more actuator devices 42a-42n that control one or more vehicle characteristics, such as the drive system 20, the transmission system 22, the steering system 24, and the braking system 26. In various embodiments, the vehicle 100 may also include interior and / or exterior vehicle equipment not specified in Fig. 1 shown, such as various doors, trunk and cabin equipment, such as air, music, lighting, touchscreen display components (as used in conjunction with navigation systems) and the like.

[0039] The data storage device 32 stores data for use in the automatic control of the vehicle 100. In various embodiments, the data storage device 32 stores defined maps of the navigable environment. In various embodiments, the defined maps can be predefined and retrieved from a remote system. For example, the defined maps can be compiled by the remote system and communicated to the vehicle 100 (wirelessly and / or via a wired connection) and stored in the data storage device 32. Route information can also be stored in the data storage device 32—that is, in a series of road segments (geographically linked to one or more of the defined maps) that together define a route that the user can travel from a starting point (e.g., the user's current location) to a destination.As can be seen, the data storage device 32 can be part of the controller 34, separate from the controller 34, or part of the controller 34 and part of a separate system.

[0040] The controller 34 includes at least one processor 44 and a computer-readable memory device or media 46. The processor 44 can be a custom-designed or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC) (e.g., a user-defined ASIC implementing a neural network), a field-programmable gate array (FPGA), an auxiliary processor among multiple processors connected to the controller 34, a semiconductor-based microprocessor (in the form of a microchip or chipset), a combination thereof, or generally any device for executing instructions. The computer-readable memory device or media 46 can include volatile and non-volatile memory in a read-only memory (ROM), a direct-access memory (RAM), and a keep-alive memory (KAM).KAM is a persistent or non-volatile memory that can be used to store various operating variables while the processor 44 is powered off. The computer-readable memory device or media 46 can be implemented using any number of known memory devices, such as PROMs (programmable read-only memory), EPROMs (electrical PROMs), EEPROMs (electrically erasable PROMs), flash memory, or any other electrical, magnetic, optical, or combined memory devices capable of storing data, some of which represent executable instructions used by the controller 34 in controlling the vehicle 100. In various embodiments, the controller 34 is configured to implement the mapping system described in detail below.

[0041] The instructions can include one or more separate programs, each comprising an ordered list of executable instructions for implementing logical functions. When executed by the processor 44, the instructions receive and process signals (e.g., sensor data) from the sensor system 28, perform logic, calculations, procedures, and / or algorithms for the automatic control of the vehicle 100's components, and generate control signals that are transmitted to the actuator system 30 to automatically control the vehicle 100's components based on the logic, calculations, procedures, and / or algorithms. Although in Fig. 1 where only one controller 34 is shown, embodiments of the vehicle 100 may include any number of controllers 34 which communicate and interact via a suitable communication medium or a combination of communication media to process the sensor signals, perform logics, calculations, procedures and / or algorithms, and generate control signals to automatically control the functions of the autonomous vehicle 100.

[0042] The communication system 36 is configured to wirelessly transmit information to and from other units 48, such as other vehicles (“V2V” communication), infrastructure (“V2I” communication), networks (“V2N” communication), pedestrians (“V2P” communication), remote transportation systems, and / or user devices. In one exemplary embodiment, the communication system 36 is a wireless communication system configured to communicate via a wireless local area network (WLAN) using the IEEE 802.11 standard or via mobile data communication. However, the scope of the present invention also includes additional or alternative communication methods, such as a dedicated short-range communication (DSRC) channel.DSRC channels refer to one-way or two-way short-range to medium-range radio communication channels specifically designed for the automotive industry and a corresponding set of protocols and standards.

[0043] According to various embodiments, the controller 34 can be an autonomous driving system (ADS) 70, as in Fig. 2 shown. This means that suitable software and / or hardware components of the controller 34 (e.g. the processor 44 and the computer-readable storage medium 46) can be used to provide an autonomous drive system 70 that is used in conjunction with the vehicle 100.

[0044] In various embodiments, the instructions of the autonomous drive system 70 can be structured according to function or system. For example, the autonomous drive system 70 can be structured as shown in Fig. Figure 2 shows a perception system 74, a positioning system 76, a path planning system 78, and a vehicle control system 80. As can be seen, the instructions can be divided into any number of systems (e.g., combined, further subdivided, etc.) in various embodiments.

[0045] In various embodiments, the perception system 74 synthesizes and processes the acquired sensor data and predicts the presence, location, classification, and / or movement of objects and features in the vehicle 100's environment. In various embodiments, the perception system 74 can incorporate information from multiple sensors (e.g., the sensor system 28), including cameras, lidars, radars, and / or any number of other sensor types. In various embodiments, the object detection and tracking system 302 can be fully or partially integrated into the perception system 74.

[0046] The positioning system 76 processes sensor data together with other data to determine the position (e.g., a local position relative to a map, an exact position relative to a road lane, vehicle direction, etc.) of the vehicle 100 in relation to its environment. As can be seen, various techniques can be used to perform this localization, such as simultaneous localization and mapping (SLAM), particle filters, Kalman filters, Bayesian filters, and the like.

[0047] The path planning system 78 processes sensor data together with other data to determine a path that the vehicle 100 should follow. The vehicle control system 80 generates control signals to control the vehicle 100 according to the determined route.

[0048] In various embodiments, the controller 34 implements machine learning techniques to support the functionality of the controller 34, such as feature recognition / classification, obstacle mitigation, route crossing, mapping, sensor integration, ground truth determination, and the like.

[0049] Fig. Figure 3 is a block diagram of the object detection and tracking system 302 in detail according to exemplary embodiments. The exemplary object detection and tracking system 302 is configured to retrieve sensor measurements 301 from the radar sensors 308, generate refined measurements 303 for each set of sensor measurements 301 by eliminating measurements that contradict a direction of travel specified by lanes, associate the refined measurements 303 with tracked objects, and provide object tracking data 305 to other vehicle systems. The exemplary object detection and tracking system 302 includes a measurement detection system 304 and a tracking and data mapping system 306. The object detection and tracking system 302 can be operated on the controller 34 of Fig. 1, implemented on a separate controller or on a combination of controllers in various embodiments.

[0050] The exemplary measurement acquisition system 304 is configured to retrieve radar measurements 301, which are performed by radar sensors 308 in the vehicle 300 at various periodic time steps. The radar measurements 301 can include position data, velocity data, signal-to-noise ratio (SNR) data, power data, etc., for one or more objects during a specific period. The exemplary measurement acquisition system 304 is configured to organize the radar measurements as time-ordered groups of radar measurements into suitable time windows. The time window in which a time-ordered cluster of radar measurements is organized is determined based on the period in which the measurements were received.

[0051] The exemplary radar measurement system 304 is further configured to build a sequence cluster of radar measurements. The sequence cluster comprises several time-ordered clusters of radar measurements corresponding to a common object. The multiple time-ordered clusters within the sequence cluster are arranged in chronological order. The sequence cluster is also configured as a sliding window of radar measurements. The sliding window of radar measurements within the sequence cluster contains a predetermined number of the most recent time windows of radar measurements. In one example, the predetermined number is five, but in other examples, the predetermined number of time windows may differ. Since a newer time window of radar measurements is added to the sequence cluster after the predetermined number of time windows is reached, the oldest time window is removed from the sequence cluster.

[0052] The Fig. Figures 4A-4B provide an exemplary representation of time-ordered clusters of radar measurements and sequence clusters of radar measurements. Fig. 4A is a diagram showing exemplary radar measurements (including position data 401 and speed data 403) in time intervals t0, t1 and t2. Fig. Figure 4B is a diagram illustrating the assignment of exemplary radar measurements in the time intervals t0, t1, and t2 to a sequence cluster 402 and a direction of movement 404 for sequence cluster 402. As illustrated, the exemplary sequence cluster 402 contains three time-ordered clusters of radar measurements. One cluster corresponds to time window t0, a second cluster corresponds to time window t1, and a third cluster corresponds to time window t2. The multiple (three) clusters in sequence cluster 402 are arranged in chronological order, and sequence cluster 402 can be arranged as a sliding window of radar measurements.

[0053] With renewed reference to Fig. 3. Exemplary Measurement System 304 is further configured to remove noise from the sequence cluster of radar measurements corresponding to the common object by removing a cluster of radar measurements from the sequence cluster that contradicts a road topology map for an area where the common object is estimated to be located. Exemplary Measurement System 304 is configured to remove noise from the sequence cluster of radar measurements by comparing the sequence cluster of radar measurements with the road topology map. The road topology map is configured to identify lanes and permissible directions of travel within the identified lanes.After comparison, the exemplary radar acquisition system 304 is configured to remove noise by removing radar measurements from the sequence cluster that indicate object movement in a direction that contradicts a permissible direction of travel in the lanes of the road topology map. The exemplary radar acquisition system 304 is further configured to output the sequence cluster of radar measurements as a new cluster of radar measurements after attempting to remove conflicting radar measurements.

[0054] Fig. Figures 4C-4D illustrate the use of road topology by an exemplary measuring system 304. Fig. 4C is a diagram that maps the superposition of tracks 406 and 408 onto sequence cluster 402. In this example, sequence cluster 402 indicates that the direction of travel 404 of the object represented by the radar measurements is in the opposite direction to the permissible direction of travel 407 for track 406 and in the same direction as the permissible direction of travel 409 for track 408. Therefore, the radar measurements in track 406 may contain noise and can be removed from sequence cluster 402 of the radar measurements.

[0055] Fig. 4D is a diagram that maps the superposition of lanes 416 and 418 onto sequence cluster 402. In this example, sequence cluster 402 indicates that the direction of travel 404 of the object represented by the radar measurements is in the same direction as the permissible direction of travel 417 and 419 for lane 416 and for lane 418. Therefore, the object represented by sequence cluster 402 from radar measurements can be located in either lane 416 or 418. In this example, the use of a road topology map alone cannot be used to reduce the noise from sequence cluster 402 of the radar measurements.

[0056] The Fig. 5A-5C provides a further exemplary illustration of the use of road topology by an exemplary measuring system 304. Fig. 5A is a diagram showing exemplary radar measurements at time intervals t0, t1, t2, t3, t4 and t5. Fig. 5B is a diagram that shows the direction of travel 502 of the radar measurements in the time intervals t0, t1, t2, t3, t4 and t5. Fig. 5C is a diagram that shows the superimposition of lanes 504 and 506 on the exemplary radar measurements. In this example, the direction of travel of the object represented by the radar measurements corresponds to the permitted direction of travel for lanes 504 and 506. Therefore, in this example, the use of a road topology map alone cannot be used to reduce the noise from the radar measurements.

[0057] With renewed reference to Fig. 3 The exemplary tracking and data mapping system 306 is configured to further reduce noise in the radar measurements, to determine whether the radar measurements of the measuring acquisition system 304 should be assigned to one or more radar tracks (each radar track comprising successive observations of the same object), to track the radar tracks with the radar measurements, and to predict future radar measurements with the radar tracks and road topology information. The exemplary tracking and data mapping system 306 includes a data mapping module 312 and a tracking module 314.

[0058] The exemplary tracking module 314 is configured to track one or more radar tracks using a separate instance of a restricted filter 316, such as a restricted Kalman filter 316, for each radar track. The exemplary tracking module 314 is configured to perform the tracking by forcing and restricting the movement of the one or more radar tracks within an aligned track in a manner prescribed by the aligned track, even if a measurement supporting movement is missing for one or more radar tracks. For example, if a radar track indicates that an object is moving within a lane (e.g., an aligned lane) and a radar measurement is not received during a time window (e.g.,(If an obstructing object prevents a radar measurement of the object from being received), the exemplary tracking module 314 is configured via the restricted Kalman filter 316 to predict the object's position and speed during the time interval in which the radar measurement was not received. This prediction is based on previous kinematics (e.g., speed, acceleration, time interval, distance traveled, etc.) with respect to the object and the restrictions (e.g., direction of travel, speed limit, turning requirements or restrictions, etc.) applied to the object's movement according to the lane rules. Furthermore, the exemplary tracking module 314 is configured to predict future object movements based on past kinematics and lane restrictions.

[0059] The exemplary data mapping module 312 is configured to determine whether a received new cluster of radar measurements corresponds to a detected object being tracked by the exemplary tracking module 314 (for example) and aligned within a track, by comparing the new cluster of radar measurements to the radar tracks aligned within a track. If a new cluster of radar measurements does not correspond to a tracked object, then the new cluster could correspond to a new object or to noise. The exemplary data mapping module 312 is further configured to attempt to map the new cluster of radar measurements to a radar track corresponding to a tracked object if the new cluster of radar measurements appears to correspond to a tracked object.

[0060] When attempting to map the new cluster of radar measurements to a radar track corresponding to a tracked object, the exemplary data mapping module 312 is configured to search aligned tracks for previous radar measurements in earlier time windows that correspond to a radar track and match the new cluster of radar measurements. The exemplary data mapping module 312 is further configured to link the new cluster of radar measurements to the identified radar track if the earlier radar measurements corresponding to the identified radar track match the new cluster of radar measurements.

[0061] The Fig. 6A and Fig. Section 6B provides examples of previous radar measurements that correspond to a radar track matching a new cluster of radar measurements. Fig. Figure 6A is a diagram depicting an exemplary cluster 602 of radar measurements taken at time steps t0, t1, t2, and t3, relative to exemplary aligned lanes. The aligned lanes from which an object represented by the exemplary cluster 602 of radar measurements could have originated include lane 604 and lane 606. The exemplary data mapping module 312 is configured to search within the aligned lanes. In this example, the exemplary data mapping module 312 would determine that radar measurements of lane 604, taken at time steps t0, t1, t2, and t3, could have originated from lane 604 and lane 606. -5 and t -6 were carried out, with (e.g., agree, the movement of an object represented by the radar measurements of track 604, which were taken at time steps t -5 and t -6 The measures taken could lead to) the exemplary block 602 of radar measurements.

[0062] Fig. Figure 6B is a diagram depicting an exemplary cluster 612 of radar measurements taken at time steps t0, t1, and t2, relative to exemplary aligned lanes. The aligned lanes from which an object represented by the exemplary cluster 612 of radar measurements could have originated include lane 614 and lane 616. The exemplary data mapping module 312 is configured to search within the aligned lanes. In this example, the exemplary data mapping module 312 would determine that radar measurements of lane 614, taken at time steps t0, t1, and t2, could have originated from lane 614 and lane 616. -4 and t -8 were carried out, with (e.g., agree, the movement of an object represented by the radar measurements of track 614, which were taken at time steps t -4 and t -8 The measures taken could lead to) the exemplary block 612 of radar measurements.

[0063] With renewed reference to Fig. 3 The exemplary tracking and data mapping system 306 is further configured to predict a future observation for the identified radar track by projecting the future observation onto a path specified by the aligned track. Fig. 7A, Fig. 7B and Fig. 7C shows examples of the projection of a future observation along a path defined by an aligned track.

[0064] Fig. Figure 7A is a diagram depicting exemplary radar measurements at time steps t0, t1, and t2 with respect to an object in lane 702. Radar measurements from a vehicle 704 can be blocked from the vehicle's view by an obscuring object 706 for part of the journey of an object represented by the exemplary radar measurements taken at time steps t0, t1, and t2 along lane 702. The exemplary tracking module 314 is configured to track the radar trail 708, which is represented by the exemplary radar measurements taken in time steps t0, t1 and t2 using a restricted filter, such as a restricted Bayesian filter, a restricted Kalman filter 316, a restricted particle filter, an unscented Kalman filter, a long-term storage (LSTM) filter, a Bayesian inference filter or the like.The exemplary tracking module 314 is configured to perform tracking by enforcing and constraining the movement of radar track 708 within the aligned track 702 in a manner prescribed by the aligned track, even if a measurement supporting movement is absent on the radar track. In this example, track 702 dictates that the object must maintain its current direction of travel. The exemplary tracking module 314 can estimate the projected movement 710 of the object and predict that the object will maintain its current direction of travel within track 702 and will be positioned around specified points within track 702 at appropriate time intervals.

[0065] Fig. Figure 7B is a diagram depicting exemplary radar measurements at time steps t0, t1, and t2 with respect to an object in lane 712. The exemplary tracking module 314 is configured to track radar track 714, which is represented by the exemplary radar measurements at time steps t0, t1, and t2 using a restricted filter 316. The exemplary tracking module 314 is configured to perform the tracking by forcing and restricting the movement of radar track 714 in a manner dictated by the aligned lane 712. In this example, lane 712 is a right-turn lane only, so the object must turn right into lane 716. The exemplary tracking module 314 can estimate the projected movement 718 of the object and predict that the object will turn right into track 716 during a reasonable time interval.

[0066] Fig. Figure 7C is a diagram that maps exemplary radar measurements at time steps t0, t1, and t2 with respect to an object in track 722. The exemplary tracking module 314 is configured to track radar track 724, which is represented by the exemplary radar measurements at time steps t0, t1, and t2 using a constrained filter 316. The exemplary tracking module 314 is configured to perform the tracking by forcing and constraining the movement of the radar track in a manner dictated by the aligned track 722. In this example, the topology of track 722 dictates that the object must turn left via track 726. The exemplary tracking module 314 can estimate the projected movement 728 of the object and predict that the object will cross track 726 during a suitable time interval.

[0067] With renewed reference to Fig. 3 The exemplary tracking and data mapping system 306 is further configured to reject the new cluster of radar measurements as noise if the earlier radar measurements corresponding to the identified radar tracks do not match the new cluster of radar measurements.

[0068] Fig. Figure 8 is a process flow diagram illustrating an exemplary process for object detection and tracking using radar measurements. The sequence of operations within the method is not limited to the sequential execution shown in the figure, but can be carried out in one or more different sequences according to the present invention. In various embodiments, the method can be executed based on one or more predefined events and / or continuously during the operation of the vehicle 100.

[0069] Exemplary process 800 involves retrieving radar measurements taken at various periodic time steps by a radar system in the vehicle (operation 802). The radar measurements can include position and velocity data for one or more objects during a specific period. The radar measurements can be retrieved directly from a radar sensor or a network interface.

[0070] Exemplary process 800 involves scheduling the radar measurements within a suitable time window (operation 804). This can involve organizing the radar measurements into appropriate time windows, where the time window in which a set of radar measurements is organized corresponds to the time span in which the set of radar measurements was performed. The radar measurements can be framed within a sliding window of radar measurements. The sliding window can consist of the most recent predetermined number of data frames. In one example, the predetermined number is five, but other predetermined numbers can also be used.

[0071] Exemplary process 800 involves building a sequence cluster of measurements (operation 806). The sequence cluster of measurements should all refer to a common object. The sequence cluster can include radar measurements corresponding to a common object across a variety of different time windows. The sequence cluster can include a sliding window of radar measurements, where the sliding window contains a predetermined number of the most recent time windows of radar measurements. In one example, the predetermined number is five, but other predetermined numbers can also be used.

[0072] Exemplary process 800 involves referencing the lane topology to remove conflicting groups of radar measurements from a sequence measurement cluster and generate a new cluster of radar measurements (Operation 808). Moving objects can be detected in a noisy radar signal at greater distances with fewer measurements by referencing the lane topology to remove conflicting groups of radar measurements from a sequence of measurements. Referencing the lane topology to remove conflicting radar measurement clusters from a sequence cluster of measurements can involve removing noise from the sequence cluster of radar measurements corresponding to the common object by removing a cluster of radar measurements from the sequence cluster of radar measurements that conflicts with a lane topology map for an area where the first object is estimated to be local.The underlying road and lane information can be used to filter and remove elements, such as noise, from the sequence cluster of measurements. Removing a cluster of radar measurements that conflicts with a road topology map may involve comparing the sequence cluster of radar measurements with a road topology map configured to identify lanes and permissible directions of travel within those lanes. Removing a cluster of radar measurements that conflicts with a road topology map may also involve removing radar measurements indicating object movement in a direction that conflicts with a permissible direction of travel within the lanes shown on the road topology map.

[0073] Exemplary Process 800 involves tracking one or more radar tracks over time, for example, using a separate instance of a restricted filter, such as a restricted Kalman filter, for each radar track (Operation 810). Each radar track comprises a sequence of observations of the same object. The tracking may involve forcing and restricting the motion of one or more radar tracks within an aligned track in a manner prescribed by the aligned track, even if a measurement supporting motion is missing in one or more radar tracks.

[0074] Exemplary process 800 involves referencing a map of detected objects aligned within a lane (operation 812). This map of detected objects aligned within a lane can be referenced to determine whether a new cluster of radar measurements can correspond to a detected object aligned within a lane. This allows objects moving within lanes to be detected earlier and their speed to be restricted to comply with lane requirements.

[0075] Exemplary process 800 involves determining whether an object has been detected (decision 814). By comparing a new cluster of radar measurements with radar tracks aligned within a lane, it can be determined whether the new cluster of radar measurements corresponds to a detected object aligned within a lane. This can enable a search for confirmation that the new cluster represents an object moving within a lane and allow a vehicle to respond quickly to objects moving in this manner.

[0076] Exemplary Process 800 involves searching a set of aligned tracks for previous measurements (Operation 816). If the comparison identifies a radar track that the new cluster could correspond to (yes, in Decision 814), the search of aligned tracks for previous radar measurements that correspond to the identified radar track and are consistent with the new cluster of radar measurements can be performed. If the comparison does not identify a radar track that the new group could correspond to (no, in Decision 814), the new group of radar measurements cannot be assigned to any of the radar tracks and can be rejected as noise.

[0077] Exemplary Process 800 involves matching measurements over time to new measurements to existing lanes (Operation 818). If the earlier radar measurements corresponding to the identified radar lane match the new cluster of radar measurements, the new cluster can be associated with the identified radar lane. This can confirm that an object is moving within a lane and allow the vehicle to predict that the object will continue to move in a manner that meets the lane requirements. If the earlier radar measurements corresponding to the identified radar lane do not match the new cluster of radar measurements, the new cluster can be rejected as noise.

[0078] The exemplary process can further involve predicting a future observation for the identified radar track by projecting the future observation onto a path predetermined by the aligned track. Predicting a future observation can occur when an obscuring object prevents the vehicle from receiving radar returns from sections of the aligned track, as in Fig. 7A illustrates this. Predicting a future observation can also include predicting that the detected object may move to a different track due to the requirements of the aligned track, as shown in Fig. Figure 7B illustrates this. For example, the aligned lane may include a turn-only lane segment, a merge-only lane segment, and / or a non-turn-only lane segment. Predicting a future observation may also include predicting that the detected object may cross into another lane due to the requirements of the aligned lane, as shown in Figure 7B. Fig. 7C illustrates.

Claims

[1] Processor-implemented method in a vehicle (100) for detecting and tracking objects using radar data, the method comprising: Retrieval of radar measurements taken at various periodic time steps by a radar system in the vehicle (100); Organizing the radar measurements as time-ordered groups of radar measurements by the processor (44) into suitable time windows, wherein the time window in which a time-ordered cluster of radar measurements is organized corresponds to the time span in which the time-ordered cluster of radar measurements was carried out; Building a sequence cluster of radar measurements by the processor (44), wherein the sequence cluster comprises several time-ordered clusters of radar measurements corresponding to a first object and the several time-ordered clusters in the sequence cluster are arranged in chronological order, wherein the sequence cluster is arranged as a sliding window of radar measurements and the sliding window includes a predetermined number of the most recent time windows of radar measurements; Removal of noise from the sequence cluster of radar measurements by the processor (44) by removing a cluster of radar measurements from the sequence cluster of radar measurements that is inconsistent with a street topology map for an area where the first object is estimated to be localized; and Output the sequence cluster of radar measurements after removing contradictory radar measurements as a new cluster of radar measurements; removing noise from the sequence cluster of radar measurements includes the following: Comparing the sequence cluster of radar measurements with the road topology map, wherein the road topology map is configured to identify lanes and permissible directions of travel within the identified lanes; and Removal of radar measurements indicating object movements in a direction that contradicts a permissible direction of travel in the lanes of the road topology map. [2] Processor-implemented method according to claim 1, further comprising: Tracking one or more radar tracks using a separate instance of a restricted Kalman filter for each radar track, wherein each radar track includes successive observations of the same object, wherein the tracking involves forcing and restricting the movement of the one or more radar tracks within an aligned track in a manner dictated by the aligned track, even if a measurement supporting movement is missing in the one or more radar tracks; Determine whether the new cluster of radar measurements can correspond to a detected object aligned within a lane by comparing the new cluster of radar measurements with the radar tracks aligned within a lane; and Do not assign the new cluster of radar measurements to one of the radar tracks if the comparison does not lead to the identification of a radar track that the new cluster can correspond to. [3] Processor-implemented method according to claim 2, further comprising: If the comparison leads to the identification of a radar track that the new cluster may correspond to, search aligned tracks for previous radar measurements that correspond to the identified radar track and are consistent with the new cluster of radar measurements; and Assigning the new cluster of radar measurements to the identified radar track if the previous radar measurements corresponding to the identified radar tracks match the new cluster of radar measurements. [4] Processor-implemented method according to claim 3, further comprising rejecting the new cluster of radar measurements as noise if the earlier radar measurements corresponding to the identified radar tracks do not match the new cluster of radar measurements.

Images