Radar detection prioritization based on downstream feedback

Abstract

Methods and systems implementing radar detection prioritization based on downstream feedback are described. As a vehicle is traveling in a travelway, radar detections are received by a processor and feedback related to a downstream function is received from another processor that is performing the downstream function. For radar detections determined to be within the travelway, a respective priority is assigned based on the location of the radar detection and the feedback related to the downstream function. The radar detections with the assigned priorities are then stored in a detection buffer along with an indication of their respective assigned priorities. In this way, the detection buffer can be optimized for the downstream function by filtering out certain detections and prioritizing others. As a result, the computational burden on the downstream function is reduced while ensuring that important detections are received by the downstream function as early as possible.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 148,973, filed February 12, 2021, pursuant to 35U.SC119(e), the disclosure of which is incorporated herein by reference in its entirety. Background Technology

[0003] Radar is commonly used in vehicles to support various driver assistance and autonomous driving functions. The radar system, upstream of these functions, generates radar detections from received radar reflections and places these detections in buffers accessible to the vehicle's functions. Downstream functions receive the detections from the buffers and perform their corresponding operations. Often, radar buffers become saturated (e.g., full) due to detections corresponding to nearby and large objects. Some of these objects (or portions thereof) may not be particularly important to downstream functions. Therefore, the buffer may be filled with unimportant detections; some important detections may have no space within the buffer and thus never be processed by downstream functions. Furthermore, some more meaningful detections may be processed later than they could otherwise be processed (e.g., because they can only be stored when space becomes available) or simply not processed at all. Summary of the Invention

[0004] The aspects described below include a method for prioritizing radar detections based on downstream feedback, executed by a processor of a vehicle. The method includes: receiving radar detections while the vehicle is traveling in a corridor, the radar detections having an associated position relative to the vehicle. The method further includes: receiving feedback from another processor of the vehicle related to a downstream function executed by that other processor while the vehicle is traveling in the corridor, and determining whether any of the radar detections is within the corridor based on the position of the radar detections and the feedback related to the downstream function. For at least one radar detection within the corridor, the method further includes: assigning a priority to the at least one radar detection, the priority depending on the position of the at least one radar detection and the feedback related to the downstream function. For at least one radar detection within the corridor, the method further includes: storing an indication of the at least one radar detection and the assigned priority in a detection buffer.

[0005] Aspects described below also include a system configured to be implemented in a vehicle for radar detection prioritization based on downstream feedback. The system includes a processor and a computer-readable storage medium including instructions that, when executed by the processor, cause the processor to receive radar detections while the vehicle is traveling in a travelway. The radar detections have associated locations relative to the vehicle. The instructions further cause the processor to receive feedback related to a downstream function performed by another processor of the vehicle from the other processor while the vehicle is traveling in the travelway, and determine whether any of the radar detections are within the travelway based on the locations of the radar detections and the feedback related to the downstream function. The instructions also cause the processor to assign a priority to at least one of the radar detections that is within the travelway, the priority depending on the location of the at least one of the radar detections and the feedback related to the downstream function. The instructions further cause the processor to store the at least one of the radar detections and an indication of the assigned priority in a detection buffer. BRIEF DESCRIPTION OF DRAWINGS

[0006] Systems and techniques for radar detection prioritization based on downstream feedback are described with reference to the following figures. Like numerals may refer to like features throughout the figures:

[0007] Figure 1 is an example illustration of an environment in which radar detection prioritization based on downstream feedback can be implemented in accordance with the techniques of this disclosure;

[0008] Figure 2 is an example illustration of a system configured to perform radar detection prioritization based on downstream feedback in accordance with the techniques of this disclosure;

[0009] Figure 3 is an example illustration of radar detection prioritization based on downstream feedback in accordance with the techniques of this disclosure;

[0010] Figure 4 is another example illustration of radar detection prioritization based on downstream feedback in accordance with the techniques of this disclosure; and

[0011] Figure 5 is an example illustration of a method of radar detection prioritization based on downstream feedback in accordance with the techniques of this disclosure. DETAILED DESCRIPTION

[0012] SUMMARY

[0013] Radar is commonly used to support various driver assistance and autonomous driving functions. Many times, radar detections (e.g., radar returns) of objects are placed in a buffer for downstream processing. The downstream functions then pull the detections from the buffer to perform various operations. As long as there is space in the buffer, detections can be added without issue. However, when the buffer is full, the radar system can wait until space is freed up (e.g., detections are pulled out by downstream functions) before adding new detections.

[0014] This can result in processing of important detections being delayed or even missed. Because storage of detections in the buffer is not influenced, the buffer can fill up with detections that are not particularly relevant to downstream operations. For example, the buffer can become saturated with detections of an upcoming bridge even though the upcoming bridge can not be relevant to downstream functions (e.g., speed control using a vehicle ahead). Thus, the buffer can not be able to store detections indicating slow moving traffic ahead that can be more relevant to downstream operations than the bridge detections. By not influencing storage of detections, processing of important detections can be delayed or simply missed. This can result in unsafe driving conditions, reduced safety, and reduced driver satisfaction.

[0015] Methods and systems are described that implement radar detection prioritization based on downstream feedback. Radar detections are received while a vehicle is traveling a travelway, the radar detections having associated locations relative to the vehicle. Feedback related to downstream functions is received from another processor while the vehicle is traveling the travelway, and it is determined whether any of the radar detections are within the travelway based on the locations and the feedback. For at least one of the radar detections that are within the travelway, a priority is assigned based on the location of the at least one of the radar detections and the feedback related to the downstream functions. The at least one of the radar detections is then stored in a detection buffer along with an indication of the assigned priority. In this way, the detection buffer can be optimized for downstream functions by filtering out certain detections and prioritizing other detections. The downstream functions can process the detections that are most important to driving safety and situational awareness without having to burden themselves with evaluating less important detections. Thus, the computational burden on the downstream functions is reduced because the receipt of important detections by the downstream functions is expedited.

[0016] Example Environment

[0017] Figure 1is an example illustration 100 of an environment in which radar detection prioritization based on downstream feedback can be implemented. The example illustration 100 shows a radar system 102 of a system (not shown) within a vehicle 104. The vehicle 104 is traveling in a travel lane 106, which can correspond to a portion of a road on which the vehicle 104 is traveling. For example, the travel lane 106 can correspond to a portion of the road in the same direction of travel as the vehicle 104 (e.g., one side of the road). The travel lane 106 can be defined by an extent 108 (e.g., a barrier, fence, guardrail, or edge of a sidewalk / concrete) on lateral edges of the travel lane 106.

[0018] As the vehicle 104 traverses the travel lane 106, radar detections (detections 110) are received. The detections 110 include information about objects 112 (or portions thereof) proximate to the vehicle 104 derived from radar reflections. For example, detection 110-1 corresponds to object 112-1 (e.g., a cloverleaf support), detection 110-2 corresponds to object 112-2 (e.g., another cloverleaf support), and detection 110-3 corresponds to object 112-3 (e.g., another vehicle). The detections 110 have associated position and motion information relative to the vehicle 104.

[0019] It should be noted that the detections 110 are merely representations of the respective objects 112 (or portions thereof). That is, the detections 110 themselves do not have position or motion data; rather, the corresponding objects 112 have position or motion data. For simplicity, the detections 110 are used herein. For example, by stating that a detection 110 is moving, it should be understood that the object (or portion thereof) to which the detection 110 corresponds is moving.

[0020] The radar system 102 uses feedback received from downstream functions to prioritize the detections 110. For example, the radar system 102 can use the feedback to determine which of the detections 110 are outside of the travel lane 106. The radar system 102 can then refrain from placing those detections 110 outside of the travel lane 106 into a detection buffer for processing by downstream functions. For detections 110 that are within the travel lane 106, the radar system 102 can prioritize the detections 110 and place them in the detection buffer for processing by downstream functions.

[0021] For example, in the example illustration 100, using conventional techniques, the detection buffer can become full or saturated with detections 110-1 and 110-2 corresponding to the overpass, as the overpass is large and close to the vehicle. In this case, the detection 110-3 corresponding to the other vehicle can not be stored within the detection buffer. In many cases, the detection of a moving object (the other vehicle) within the travel lane 106 can be more important than a static object (the overpass) outside the travel lane 106. Thus, using the techniques described herein, the detection buffer can be optimized for relevance (e.g., by storing the detection 110-3 while not storing the detections 110-1 and 110-2), while reducing the computational burden on downstream functions. Doing so can enable certain detections 110 to be processed earlier, thereby improving safety and driver satisfaction.

[0022] Example System

[0023] Figure 2 is an example illustration 200 of a system 202 in which radar detection prioritization based on downstream feedback can be implemented. The system 202 is configured to be disposed in the vehicle 104. Although the vehicle 104 is illustrated as a car, the vehicle 104 can include any vehicle (e.g., a truck, a bus, a boat, an airplane, etc.) without departing from the scope of the present disclosure. As shown below, the system 202 includes the radar system 102 and an environment perception system 204. The radar system 102 and the environment perception system 204 are illustrated as having respective processors 206 (e.g., 206-1 and 206-2) and respective computer-readable storage media 208 (e.g., 208-1 and 208-2). In some implementations, the radar system 102 and the environment perception system 204 can share the processors 206 and / or the computer-readable storage media 208.

[0024] The processors 206 (e.g., application processors, microprocessors, digital signal processors (DSPs), or controllers) execute instructions 210 (e.g., code) stored in the computer-readable storage media 208 (e.g., a non-transitory storage device such as a hard disk drive, SSD, flash memory, read only memory (ROM), EPROM, or EEPROM) to cause the radar system 102 and the environment perception system 204 to perform the techniques described herein.

[0025] Instructions 210 cause radar system 102 and environment perception system 204 to operate on (e.g., create, receive, modify, delete, send, or display) data 212 (e.g., application data, module data, sensor data, or I / O data). Although shown as within respective computer-readable storage media 208, portions of data 212 can be within random access memory (RAM) or cache (not shown) of radar system 102 and environment perception system 204. Moreover, instructions 210 and / or data 212 can be located remote from radar system 102 and and environment perception system 204.

[0026] Environment perception system 204 is configured to provide downstream feedback to radar system 102 for radar detection prioritization. Radar system 102 is configured to receive the downstream feedback from environment perception system 204 and use it to perform radar detection prioritization. To do so, radar system 102 can include or interface with radar sensors (not shown), and environment perception system 204 can include or interface with other types of sensors (e.g., lidar, imager, laser, or GNSS) not shown.

[0027] Example data flow

[0028] Figure 3 is an example illustration 300 of radar detection prioritization based on downstream feedback. Example illustration 300 shows aspects of radar system 102 and environment perception system 204. Example illustration 300 further depicts data flow between radar system 102 and environment perception system 204.

[0029] Radar system 102 includes prioritization module 302, which receives detections 110 and coefficients and constants 304 (e.g., downstream feedback) from environment perception system 204. Detections 110 are generally unfiltered, not prioritized, and can be determined by radar system 102. Detections 110 have an associated location (e.g., Cartesian coordinates) and velocity (e.g., rate of change of distance, relative velocity, or absolute velocity) relative to vehicle 104.

[0030] Prioritization module 302 uses the location, as well as coefficients and constants 304, to determine prioritized detections 306, as will be discussed further Figure 4 with respect. Prioritized detections 306 are any of detections 110 that have been assigned a priority greater than zero. Prioritized detections 306 are placed into detections buffer 308 by prioritization module 302.

[0031] Depending on the application, the detection buffer 308 can have various structures or types, including derivatives of buffer types. As just one example, the detection buffer 308 can function as a first-in, first-out (FIFO) buffer, and overwrite the oldest detections before overwriting any more recent detections to make room.

[0032] In some implementations, the prioritization module 302 can prioritize the detections 110 only in response to determining that the detection buffer 308 is full. For example, if there is room in the detection buffer 308, the prioritization module 302 can not prioritize the detections or place all of the detections 110 (e.g., including the prioritized detections 306 and detections 110 that have been assigned a priority of zero) into the detection buffer 308. Further, at 310, the prioritization module 302 can deprioritize the prioritized detections 306 from within the detection buffer 308. For example, the prioritization module 302 can deprioritize from time to time (e.g., periodically or in response to a hardware or software interrupt) to reestablish a baseline for generating feedback. Doing so periodically (e.g., at regular intervals) keeps the detections 110 continuous.

[0033] The environment perception system 204 is a downstream module (e.g., downstream function) of the radar system 102. That is, the environment perception system 204 receives or pulls the prioritized detections 306 from the detection buffer 308. The environment perception system 204 uses the prioritized detections 306 along with other sensor data 312 (e.g., lidar, camera, sonar, GNSS, or map data) to implement a fused road model 314, a radar side obstacle model 316, and a visual road edge model 318. The environment perception system 204 can contain more models / functions than shown / discussed herein. The models / functions not shown can provide other functionality, which can or can not contribute to the coefficients and constants 304 for prioritization of the detections 110.

[0034] The fused road model 314 uses the other sensor data 312 (e.g., map data) and / or the prioritized detections 306 to determine a polynomial representation of the travel lane 106 of the vehicle 104 in Cartesian coordinates according to Equation 1:

[0035]

[0036] where x and y are Cartesian coordinates (e.g., x is left-right and y is back-front of the vehicle 104 as it travels along the travel lane 106). The coefficients ci, c2, and c3 correspond to the heading, curvature, and curvature rate, respectively, of the travel lane 106, and the constant corresponds to the extent 108 of the travel lane 106 (e.g., the edge of the road).

[0037] The radar side obstacle model 316 uses the prioritized detections 306 to determine constants The visual road edge model 318 uses other sensor data 312 (e.g., camera data) to determine constants

[0038] The constants and Two or more of the coefficients ci, c2, and c3 and the constants

[0039] The coefficients ci, c2, and c3 and the constants and The coefficients and constants 304 are fed back to the prioritization module 302 for use in prioritizing the detections 110. It should be noted that more or fewer of the coefficients and constants 304 described can be fed back or returned to the prioritization module 302. For example, not all constants can be available at any given time; thus, only one or two can be fed back. Furthermore, the three coefficients indicate a third order polynomial representation of the travel lane 106. In some implementations, a higher or lower order polynomial can be used. In such a case, more or fewer coefficients, respectively, would be fed back.

[0040] By using downstream feedback (e.g., the coefficients and constant 304 from the environment perception system 204), the radar system 102 can store more relevant and prioritized detections (e.g., prioritized detections 306) in the detection buffer 308. In doing so, downstream functions (e.g., the environment perception system 204) can be computationally optimized (e.g., by not acting on every detection 110) and able to act on relevant detections (e.g., prioritized detections 306) earlier. Moreover, by using only coefficients and at least one constant (e.g., 4-6 numbers) as feedback, the computational and storage burden on the environment perception system 204 and the radar system 102 is minimized.

[0041] Figure 4 is an example illustration 400 of radar detection prioritization based on downstream feedback. The example illustration 400 illustrates a technique performed by the prioritization module 302 to generate prioritized detections 306 for storage within the detection buffer 308.

[0042] As discussed above, the prioritization module 302 receives detections 110 as well as coefficients and constant 304. The following operations are performed on each of the detections 110 (or on each of a subset of the detections 110).

[0043] At 402, the position of the detections 110 is converted from Cartesian coordinates to Frenet coordinates 404 according to Equations 2 and 3.

[0044]

[0045]

[0046] where s and l are the Frenet coordinates 404 of the detections 110 (e.g., l is left-right of the vehicle 104 and s is back-front of the vehicle 104), x is the Cartesian coordinates of the detections 110, and the coefficients and constant are those described above.

[0047] At 406, the priority of the detections 110 is determined. To do so, the prioritization module 302 determines whether the detections 110 are within the travel lane 106 at decision 408. Since the constant received from the environment perception system 204 corresponds to the side range of the road, the prioritization module 302 compares the l coordinate of the detections 110 to the constant and A comparison is made. Also, not all of the constants can be received. If the l coordinate is outside the minimum value of the constant (e.g., if l is positive and greater than the minimum value of the constant, and if l is negative and less than the negative minimum value of the constant), then it is determined that the detection 110 is not within the travelway 106. In that case, the detection 110 is assigned a zero priority at 410. Thus, the detection 110 does not become a prioritized detection 306.

[0048] However, if it is determined at 408 that the detection 110 is within the travelway 106, then the process proceeds to decision 412. At 412, the prioritization module 302 determines whether the detection 110 is moving. To do so, the prioritization module 302 can compare the rate of change of distance and azimuth of the detection 110 to the speed of the vehicle 104. In some implementations, the detection 110 can already have an associated relative speed, absolute speed, and / or direction relative to the vehicle 104. For example, if the detection 110 already has an associated absolute speed, then the prioritization module 302 can simply determine whether it is above zero (e.g., moving positive).

[0049] If it is determined at 412 that the detection is not moving (e.g., it is stationary), then the prioritization module 302 can assign a priority to the detection 110 at 414 based on Equation 4.

[0050]

[0051] where p is the priority of the detection 110, C is a calibration constant, l is the Frechet coordinate 404 of the detection 110, and a minor is the sensor limit in the l direction.

[0052] If it is determined at 412 that the detection is moving, then the prioritization module 302 can assign a priority to the detection 110 at 416 based on Equation 5.

[0053]

[0054] where A and B are calibration constants, where A + B = 1 and A, B e [0 1], s is the Frechet coordinate 404 of the detection 110, and a major is the sensor limit in the s direction.

[0055] Accordingly, the prioritization module 302 can assign a priority of zero to detections 110 that are outside of the travel lane 106, assign a priority based on Equation 4 to detections 110 that are within the travel lane 106 and stationary, and assign a priority based on Equation 5 to detections 110 that are within the travel lane 106 and moving. As described above, detections with a priority greater than zero (e.g., prioritized detections 306) can then be stored within the detections buffer 308. By using the coefficients and constants 304, the prioritization module 302 can filter detections 110 (e.g., those with a priority of zero) and generate prioritized detections 306 that are stored into the detections buffer 308 in a computationally efficient manner. Doing so enables the environment perception system 204 to access an optimized detections buffer with the most relevant detections.

[0056] Example method

[0057] Figure 5 is an example illustration 500 of a method for radar detection prioritization based on downstream feedback. The example method 500 can be implemented with the previously described examples, such as the example illustrations 100, 300, and 400, and the system 202. Operations 502-510 can be performed by one or more entities (e.g., the radar system 102 or other modules or components of the system 202). The order in which the operations are illustrated and / or described is not intended to be a limitation, and any number or combination of the operations can be combined in any order to implement the method of the example illustration 500 or alternative methods.

[0058] The example illustration 500 generally begins, at 502, with receiving, by a processor of a vehicle, a radar detection when the vehicle is traveling in a travel lane. The radar detection corresponds to an object (or portion thereof) that is proximate to the vehicle when the vehicle is traveling in the travel lane and has a corresponding location relative to the vehicle. For example, the processor 206-1 of the radar system 102 can receive a detection 110 that corresponds to the object 112.

[0059] At 504, receiving, from another processor of the vehicle that is performing a downstream function, feedback related to the downstream function when the vehicle is traveling in the travel lane. For example, the processor 206-1 of the radar system 102 can receive the coefficients and constants 304 from the processor 206-2 of the environment perception system 204.

[0060] At 506, a determination is made as to whether any of the radar detections are within the travel lane. As described above, the radar detections have characteristics indicative of their respective objects. Thus, the radar detections themselves can not be within the travel lane; rather, their respective objects (or portions thereof) can be within the travel lane. For example, if desired, the prioritization module 302 can convert the coordinates of the locations (e.g., at 402) and compare the locations of the detections 110 to the coefficients and constants in the constants 304 (e.g., at 408) to determine whether the respective detections 110 are within the travel lane 106.

[0061] At 508, for radar detections that are within the travel lane, a priority is assigned based on the location and the feedback. For example, if the respective detection 110 is stationary, then at 414, or if the respective detection 110 is moving, then at 416, the prioritization module 302 can prioritize the detections 110 that are within the travel lane 106.

[0062] At 510, the radar detections that are within the travel lane are stored in a detection buffer along with their respective priorities. For example, the prioritization module 302 can store the prioritized detections 306 in the detection buffer 308.

[0063] By performing the above actions, the detection buffer is optimized while reducing the computational burden on the prioritization function and downstream functions. As a result, important detections can be identified more quickly, thereby improving safety and driver satisfaction.

[0064] Examples

[0065] Example 1 : A method comprising: receiving, by a processor of a vehicle, radar detections as the vehicle is traveling in a travel lane, the radar detections having associated locations relative to the vehicle; receiving, from another processor of the vehicle, feedback related to a downstream function performed by the other processor as the vehicle is traveling in the travel lane; determining, based on the locations of the radar detections and the feedback related to the downstream function, whether any of the radar detections are within the travel lane; and for at least one of the radar detections that is within the travel lane: assigning, by the processor, a priority to the at least one of the radar detections, the priority depending on the location of the at least one of the radar detections and the feedback related to the downstream function; and storing, by the processor, the at least one of the radar detections and an indication of the assigned priority in a detection buffer.

[0066] Example 2: The method of example 1, further comprising: determining, by the processor, whether at least one of the radar detections is stationary, wherein assigning the priority to at least one of the radar detections is further based on whether at least one of the radar detections is stationary.

[0067] Example 3: The method of example 1 or 2, wherein assigning the priority to at least one of the radar detections is further based on one or more radar sensor limitations or one or more calibration constants.

[0068] Example 4: The method of any preceding example, wherein the travel lane comprises a portion of a roadway corresponding to a direction in which the vehicle is traveling.

[0069] Example 5: The method of any preceding example, further comprising providing, by the processor, at least one of the radar detections to a downstream function.

[0070] Example 6: The method of any preceding example, further comprising, for at least one of the radar detections that is not within the travel lane, refraining from storing the at least one of the radar detections that is not within the travel lane in the detection buffer.

[0071] Example 7: The method of any preceding example, further comprising: removing, by the processor, the indication of the assigned priority from at least one of the radar detections in the detection buffer after a predetermined time.

[0072] Example 8: The method of any preceding example, wherein: wherein the position is received in Cartesian coordinates; further comprising: converting, by the processor, the Cartesian coordinates to Frechet coordinates based on the feedback; and wherein assigning the priority to at least one of the radar detections is further based on the Frechet coordinates of at least one of the radar detections.

[0073] Example 9: The method of example 8, wherein the feedback comprises a plurality of coefficients and one or more constants.

[0074] Example 10: The method of example 9, wherein: the plurality of coefficients correspond to a polynomial representation of the travel lane; the one or more constants correspond to lateral offsets that define a range of the travel lane; the coefficients are used to convert the Cartesian coordinates to the Frechet coordinates; and the constants are used to determine whether any of the radar detections are within the travel lane.

[0075] Example 11: A system configured to be implemented in a vehicle, the system comprising: a processor; a computer-readable storage medium comprising instructions that, when executed by the processor, cause the processor to: receive radar detections as the vehicle is traveling in a travel lane, the radar detections having associated locations relative to the vehicle; receive, from another processor of the vehicle, feedback related to a downstream function performed by the other processor as the vehicle is traveling in the travel lane; determine, based on the locations of the radar detections and the feedback related to the downstream function, whether any of the radar detections are within the travel lane; and for at least one of the radar detections that is within the travel lane: assign a priority to the at least one of the radar detections, the priority depending on the location of the at least one of the radar detections and the feedback related to the downstream function; and store, in a detection buffer, the at least one of the radar detections and an indication of the assigned priority.

[0076] Example 12: The system of Example 11, wherein: the instructions further cause the processor to: determine whether at least one of the radar detections is stationary; and assign the priority to the at least one of the radar detections is further based on whether the at least one of the radar detections is stationary.

[0077] Example 13: The system of Example 11 or 12, wherein assigning the priority to the at least one of the radar detections is further based on one or more radar sensor limitations or one or more calibration constants.

[0078] Example 14: The system of any one of Examples 11-13, wherein the travel lane comprises a portion of a roadway corresponding to a direction in which the vehicle is traveling.

[0079] Example 15: The system of any one of Examples 11-14, wherein the instructions further cause the processor to: provide the at least one of the radar detections to the downstream function.

[0080] Example 16: The system of any one of Examples 11-15, wherein the instructions further cause the processor to: for at least one of the radar detections that is not within the travel lane, refrain from storing the at least one of the radar detections that is not within the travel lane in the detection buffer.

[0081] Example 17: The system of any one of Examples 11-16, wherein the instructions further cause the processor to: remove, after a predetermined time, the indication of the assigned priority from the at least one of the radar detections in the detection buffer.

[0082] Example 18: The system of any of examples 11-17, wherein: the position is received in Cartesian coordinates; the instructions further cause the processor to convert the Cartesian coordinates to Frechet coordinates based on the feedback; and assigning the priority to the at least one of the radar detections is further based on the Frechet coordinates of the at least one of the radar detections.

[0083] Example 19: The system of example 18, wherein the feedback comprises a plurality of coefficients and one or more constants.

[0084] Example 20: The system of example 19, wherein: the plurality of coefficients correspond to a polynomial representation of the travel lane; the one or more constants correspond to lateral offsets that define a range of the travel lane; the coefficients are used to convert the Cartesian coordinates to the Frechet coordinates; and the constants are used to determine whether any of the radar detections are within the travel lane.

[0085] Example 21 : A system comprising means for performing any of examples 1-10.

[0086] Example 22: A system comprising at least one processor configured to perform any of examples 1-10.

[0087] Example 23: A computer-readable storage medium comprising instructions that, when executed by at least one processor, cause the processor to perform any of examples 1-10.

[0088] Example 24: The method of any of examples 1-10, further comprising: changing a function, operation, or control of the vehicle based on the at least one of the radar detections.

[0089] Although various implementations of radar detection prioritization based on downstream feedback have been described in language specific to certain features and / or methods, the subject of the appended claims is not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed as example implementations of radar detection prioritization based on downstream feedback. Further, although various examples have been described above with particular features, it should be understood that the particular features of one example are not necessarily used only in that example. Rather, any of the features described above and / or depicted in the drawings can be combined with any of the examples, in addition to or in lieu of other features of the examples.

Claims

1. A method comprising: receiving, by a processor of a vehicle, radar detections while the vehicle is traveling in a travel lane, the radar detections having associated positions relative to the vehicle; receiving, from another processor of the vehicle, feedback related to a downstream function performed by the other processor while the vehicle is traveling in the travel lane; determining, based on the positions of the radar detections and the feedback related to the downstream function, whether any of the radar detections are within the travel lane; and for at least one of the radar detections that is within the travel lane: assigning, by the processor, a priority to the respective radar detection based on at least one Frenet coordinate of the respective radar detection; and storing, by the processor, the respective radar detection and an indication of the assigned priority in a detection buffer; and providing the at least one of the radar detections and the respective indication of the assigned priority to the downstream function. further comprising: determining, by the processor, whether the respective radar detection is stationary, wherein assigning the priority to the respective radar detection is further based on whether the respective radar detection is stationary.

2. The method of claim 1, wherein, assigning the priority to the respective radar detection is further based on one or more radar sensor limitations or one or more calibration constants. the travel lane comprises a portion of a roadway corresponding to a direction in which the vehicle is traveling. further comprising, for at least one of the radar detections that is not within the travel lane, refraining from storing the at least one of the radar detections that is not within the travel lane in the detection buffer.

3. The method of claim 1, wherein, further comprising:

4. The method of claim 1, wherein, removing, by the processor, the indication of the assigned priority from the radar detections in the detection buffer after a predetermined time.

5. The method of claim 1, wherein, 7. The method of claim 1:

6. The method of claim 1, wherein, wherein the positions are received in Cartesian coordinates; converting, by the processor, the Cartesian coordinates to Frenet coordinates based on the feedback. the feedback comprises a plurality of coefficients and one or more constants.

9. The method of claim 8, wherein: Further comprising: the plurality of coefficients correspond to a polynomial representation of the travel lane; 8. The method of claim 7, wherein, the one or more constants correspond to lateral offsets that define a range of the travel lane; the coefficients are used to convert the Cartesian coordinates to the Frenet coordinates; and the constants are used to determine whether any of the radar detections are within the travel lane.

10. A system configured to be implemented in a vehicle, the system comprising: a processor; a computer-readable storage medium comprising instructions that, when executed by the processor, cause the processor to: receive radar detections while the vehicle is traveling in a travel lane, the radar detections having associated positions relative to the vehicle; ​ ​ ​ while the vehicle is driving in the driving lane, receiving feedback from another processor of the vehicle related to a downstream function performed by the other processor; based on the positions of the radar detections and the feedback related to the downstream function, determining whether any of the radar detections are within the driving lane; and for at least one of the radar detections that is within the driving lane: assigning a priority to the respective radar detection based on at least one Frechet coordinate of the respective radar detection; and storing the respective radar detection and an indication of the assigned priority in a detection buffer; and providing the at least one of the radar detections and the respective indication of the assigned priority to the downstream function.

11. The system of claim 10, wherein: the instructions further cause the processor to determine whether the respective radar detection is stationary; and assigning the priority to the respective radar detection is further based on whether the respective radar detection is stationary.

12. The system of claim 10, wherein, assigning the priority to the respective radar detection is further based on one or more radar sensor limitations or one or more calibration constants.

13. The system of claim 10, wherein, the driving lane comprises a portion of a road corresponding to a direction in which the vehicle is driving.

14. The system of claim 10, wherein, the instructions further cause the processor to, for at least one of the radar detections that is not within the driving lane, refrain from storing the at least one of the radar detections that is not within the driving lane in the detection buffer.

15. The system of claim 10, wherein, the instructions further cause the processor to, after a predetermined time, remove the indication of the assigned priority from the radar detections in the detection buffer.

16. The system of claim 10, wherein: the positions are received in Cartesian coordinates; the instructions further cause the processor to convert the Cartesian coordinates to Frechet coordinates based on the feedback.

17. The system of claim 16, wherein, the feedback comprises a plurality of coefficients and one or more constants.

18. The system of claim 17, wherein: the plurality of coefficients correspond to a polynomial representation of the driving lane; the one or more constants correspond to lateral offsets that define a range of the driving lane; the coefficients are used to convert the Cartesian coordinates to the Frechet coordinates; and the constants are used to determine whether any of the radar detections are within the driving lane.

Images