Method for reducing computing time in the calculation of radar information

By restricting the detection range using environmental information, the computational burden for high-resolution radar calculations is reduced, addressing the high resource demands of existing systems and lowering costs.

DE102017123904B4Active Publication Date: 2025-10-02AUMOVIO AUTONOMOUS MOBILITY GERMANY GMBH +1
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
DE102017123904
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2017-10-13
Publication Date
2025-10-02
Estimated Expiration
2037-10-13

AI Technical Summary

Technical Problem

High computational effort is required for calculating high-resolution radar information, particularly in synthetic aperture radar systems, leading to high system costs and resource demands.

Method used

Restrict the detection range by using environmental information to determine relevant areas for high-resolution radar calculations, employing a combination of radar and additional sensors or stored data to minimize computational resources.

Benefits of technology

Substantially reduces computational effort and resources needed for high-resolution radar information processing, lowering system costs while maintaining effective radar performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Method for reducing computing time when calculating radar information in a moving vehicle, comprising the following steps: - detecting or providing environmental information in the local area (A, A') detectable by the radar system; - Determining a detection area relevant for detection by the radar system based on the environmental information detected or provided; - Calculating the radar information in a locally restricted area based on the determined relevant detection range. The radar information is calculated in the radar system according to a synthetic aperture radar (SAR) algorithm, and the radar system is designed to acquire radar information based on chronologically successive individual measurements. The method further comprises: based on the determined relevant detection range, at least some of the information from individual measurements taken at different locations is disregarded when calculating the radar information.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method and a radar system in which the computing time for calculating radar information is reduced with the aid of environmental information.

[0002] Radar technologies are known from the state of the art by which high-resolution radar information, also referred to below as radar images, can be obtained. Such high-resolution radar images are particularly relevant for autonomous vehicle driving. Synthetic aperture radar (SAR), in particular, enables high spatial resolution. The calculation of high-resolution radar information, especially in SAR radar systems, requires a very high computational effort. In addition to highly precise knowledge of the trajectory of the vehicle on which the radar system is located, each raster area or pixel of the radar image must be calculated taking into account multiple ramp-like or pulse-like radar signals. Therefore, the system requirements for a radar system with high spatial resolution and real-time or quasi-real-time requirements are very high, particularly with regard to the size of the system memory and processor power.

[0003] The document DE 10 2013 207 904 A1 discloses a method for providing an efficient environment map for a vehicle.

[0004] The document DE 10 2013 018 315 A1 discloses a method for providing an environment model with an adaptive grid.

[0005] The document US 2008 / 0272955 A1 discloses a radar system with an active lens to adjust the field of view.

[0006] The document EP 3 151 035 A1 discloses a method for operating radar sensors in a motor vehicle.

[0007] The document US 2016 / 0003936 A1 discloses a target detection device with a radar sensor having a plurality of transmit and receive channels.

[0008] The document DE 10 2012 020 850 A1 discloses a method for object detection using a radar sensor having a plurality of receiving channels.

[0009] Based on this, it is the object of the invention to provide a method by means of which the computational effort for calculating high-resolution radar information can be reduced and thus the costs of the radar system can be reduced.

[0010] This object is achieved by a method having the features of independent patent claim 1. Preferred embodiments are the subject of the dependent claims. A radar system for reducing the computational effort in calculating radar information is the subject of independent patent claim 12.

[0011] According to a first aspect, the invention relates to a method for reducing computing time when calculating radar information in a moving vehicle. Environmental information is initially acquired, at least in the local area detectable by the radar system. Alternatively, the environmental information can also be provided, for example, in a memory unit of the motor vehicle or by downloading it from a server via a data connection.

[0012] Subsequently, based on the acquired or provided environmental information, a detection area relevant for detection by the radar system is determined. In particular, the environmental information can be evaluated to determine whether there are areas within the area detectable by the radar system from which no or essentially no back reflections are expected. In other words, the environmental information can be used, for example, to detect open spaces within the area detectable by the radar system. This allows the theoretically detectable area to be restricted to a smaller detection area relevant to the expected reflections.

[0013] After defining the relevant detection area, radar information is determined in a locally restricted area based on the determined relevant detection area.

[0014] The essential advantage of the method according to the invention is that by limiting the local area in which the radar information is calculated, the computational effort can be significantly reduced and thus the computer resources required for the calculation (processor power, memory, etc.) can be significantly reduced.

[0015] According to one embodiment, the environmental information is provided as map information or as map-like information. This map information or map-like information can be stored in a memory unit in the vehicle or downloaded via a data connection from a stationary server remote from the vehicle. This allows the environmental information to be provided in a technically simple manner with minimal effort.

[0016] According to one embodiment, the environmental information is at least partially acquired by at least one sensor system independent of the radar system. For example, a further sensor system can be provided in the vehicle, by means of which environmental information is acquired in real time or quasi-real time. The information provided by this further sensor system or derived therefrom provides an indication of the areas of the space detectable by the radar system in which significant reflections are to be expected or the area in which open spaces are located. Based on this, the detection range can be restricted.

[0017] According to one embodiment, the additional sensor system comprises a camera, a laser scanner, or another sensor suitable for capturing environmental information. This allows environmental information to be obtained with comparatively low computational effort (relative to the computational effort required to calculate the high-resolution radar information).

[0018] According to one embodiment, the environmental information is at least partially acquired by the radar system itself. This reduces the technical complexity of the equipment required to acquire the environmental information, as no separate, additional sensor system is required.

[0019] According to one exemplary embodiment, the radar system is operated in an operating mode with minimal computational effort to acquire environmental information. In other words, the radar system can be operated in a first operating mode, in which high-resolution radar information is calculated with a high computational effort, and a second operating mode, in which radar information with a lower resolution than the first operating mode is determined with less computational effort. The radar system can also be operated at least partially in parallel in both operating modes. However, the radar information acquired in the second operating mode is sufficient to be able to determine in which areas in the surroundings of the vehicle high-resolution radar information should be generated. In this case, one and the same set of received radar signals can be used in each of the respective operating modes to calculate the high-resolution radar information.lower-resolution radar information. In other words, the radar information obtained from the different operating modes is based on the same set of received radar signals, which have simply been post-processed differently. The high-resolution radar information can be calculated, at least in part, in parallel with the lower-resolution radar information.

[0020] Alternatively, the radar system can be switched intermittently between the first and second operating modes. This ensures that current environmental information is obtained at regular or irregular intervals, which can then be used to determine the relevant detection area.

[0021] Preferably, the time intervals during which the radar system operates in the first operating mode, i.e., the time interval between two operating cycles in the second operating mode, are selected depending on the detected ambient information and / or the vehicle speed. This allows the switching behavior to be adapted to the temporal changes in the ambient information, depending on the situation.

[0022] According to one embodiment, the acquired or provided environmental information is evaluated with regard to local areas relevant for the acquisition of radar information. For example, the environmental information can be analyzed to determine which local areas contain objects that are likely to cause significant reflections of the radar signals. This allows conclusions to be drawn from the environmental information as to how the theoretically possible detection range can be restricted to a relevant detection range without losing essential information in the high-resolution radar information.

[0023] According to one embodiment, the areas relevant for detecting radar information include at least one object with a size above an object size threshold. An “object size threshold” is understood to be a threshold value describing an object in terms of its size. “Size” refers to a geometric quantity (e.g., the size of an object on a map, in an image, etc.) but also an electrically measurable quantity (e.g., the portion of a radar signal reflected by the object, radar cross section (RCS)). The object size threshold thus specifies a limit value by which local areas are classified as relevant or irrelevant for high-resolution calculation of radar information.

[0024] According to one embodiment, the areas relevant for detecting radar information are defined in the reduced-computational effort operating mode based on detected radar information that lies above a threshold. The threshold indicates, for example, the portion of a radar signal reflected by the object, the radar reflecting area, etc. Thus, based on the threshold, the relevant detection area can be defined as the area in which radar information above the threshold was detected.

[0025] According to one embodiment, the radar system is configured to acquire radar information based on temporally successive individual measurements. Based on the determined relevant detection range, individual measurements performed by the radar system at different locations are at least partially omitted or skipped, or the information from these individual measurements is disregarded when calculating the radar information. In particular, when calculating radar information according to the SAR algorithm, radar information acquired at different locations through individual measurements is linked to achieve a synthetic increase in the antenna aperture of the radar sensor.Due to the local restriction of the detection area based on the collected or provided environmental information, individual measurements may become superfluous, so that by skipping or ignoring them, the computational effort can be further reduced.

[0026] According to one embodiment, the radar information is calculated in the radar system according to a synthetic aperture radar (SAR) algorithm. In other words, the radar system is a system operating according to the SAR principle, in which the radar information is calculated using a SAR algorithm. A variety of algorithms for SAR processing are known, for example, so-called chirp scaling, frequency scaling, or back-projection. The back-projection algorithm for calculating the radar information, in particular, is very computationally intensive, so that in this case, a significant reduction in computational effort is achieved by limiting the detection range to a relevant area.

[0027] According to one embodiment, the computationally-reduced operating mode of the radar system acquires radar information based on a digital beamforming (DBF) principle. In other words, the radar system is configured, for example, to operate the radar sensor not only according to the SAR principle, but also, in the second, computationally-reduced operating mode, according to a receiver architecture that enables digital shaping of the antenna pattern, for example, with multiple antenna main lobes oriented in different spatial directions. Based on the DBF principle, environmental information can be determined with low computational effort compared to the SAR principle.

[0028] According to a further aspect, the invention relates to a computer program product for reducing computing time when calculating radar information in a moving vehicle, wherein the computer program product comprises a computer-readable storage medium with program instructions, wherein the program instructions are executable by a processor to cause the processor to execute a method according to one of the preceding embodiments.

[0029] According to a further aspect, the invention relates to a radar system for calculating radar information in a moving vehicle. The radar system is designed to: - to receive environmental information relating to a local area detectable by the radar system; - to determine a detection area relevant for the radar system based on the environmental information received; and - To calculate radar information in a locally restricted area based on the determined relevant detection range.

[0030] “Environmental information” in the sense of the present invention is understood to mean any information that enables an evaluation of a spatial area as to whether or not it is relevant for calculating high-resolution radar information.

[0031] The term “detection area” in the sense of the present invention is understood to mean the spatial area in which radar information can be detected by the radar system.

[0032] For the purposes of the present invention, "relevant detection range" is understood to mean the spatial area in which relevant radar information can be detected by the radar system, for example, due to back reflections above a defined portion of the transmitted radar signal. The relevant detection range can, in particular, also include open spaces in which no or essentially no back reflections of the radar signal occur, but which are delimited by barriers at the front, rear, and / or sides. These can, in particular, be parking spaces or the like.

[0033] The terms “approximately”, “essentially” or “about” mean, in the sense of the invention, deviations from the exact value by + / - 10%, preferably by + / - 5% and / or deviations in the form of changes that are insignificant for the function.

[0034] Further developments, advantages, and possible applications of the invention will become apparent from the following description of exemplary embodiments and from the figures. All described and / or illustrated features, individually or in any combination, are fundamentally part of the invention, regardless of their summary in the claims or their reference back to them. The content of the claims is also incorporated into the description.

[0035] The invention is explained in more detail below with reference to exemplary embodiments and the figures. They show: Fig. 1 shows, by way of example and schematically, a process flow for calculating a radar image based on a SAR algorithm without spatial restriction of the detection range; Fig. 2 shows, by way of example and schematically, a method sequence for calculating a radar image based on a SAR algorithm with a spatial restriction of the detection range based on environmental information obtained by means of the radar system; Fig. 3 shows, by way of example and schematically, a method sequence for calculating a radar image based on a SAR algorithm with a spatial restriction of the detection area based on environmental information acquired by other sensors or provided as data; and Fig. 4 shows an example block diagram illustrating the method for reducing computing time when calculating radar information.

[0036] Fig. 1 shows an example and schematically a process flow for calculating a radar image based on a SAR algorithm without taking environmental information into account.

[0037] A vehicle with a radar system having a radar sensor 1 is moved along a trajectory from a first position to a second position and transmits radar signals to positions #0 to #L at different times. Based on these radar signals transmitted at different positions, represented by the areas A, A', the radar system receives reflected portions of this transmitted radar signal. In other words, multiple measurements are taken at different positions of the vehicle in space and thus at different relative positions of the vehicle relative to the object O at which reflections occur. These measurements are referred to below as individual measurements. Based on the received reflected portions of the radar signal, a radar image is calculated by a processor of the radar system.

[0038] To calculate the radar image using the SAR algorithm, radar information is calculated for the entire detection area, which is then displayed as a radar image RB. Assuming that the radar image comprises N × M pixels, for which radar information must be calculated based on L individual measurements, the effort increases at least quadratically, but especially cubically, with the size of the area to be detected, assuming that an increase in the number of pixels (N and M) results in an increase in the required number of individual measurements.

[0039] In particular, very high-resolution SAR algorithms, which demonstrate a high degree of flexibility in calculating radar information and which demonstrate a high degree of stability even when cornering, such as the back-projection algorithm, already have a high computational effort, so that the calculation of the radar image requires high computing resources for large detection ranges.

[0040] Fig. 2 shows an example and schematically a process flow for calculating a radar image based on a SAR algorithm, in which the computational effort is reduced by restricting the detection range based on environmental information, whereby the environmental information is obtained by means of the radar sensor itself.

[0041] In this embodiment, radar sensor 1 is again moved along a trajectory from a first position to a second position, for example, by the vehicle moving forward. Radar signals are transmitted at different positions and their reflected components are subsequently received (positions #0 to #L). The goal of the sequentially transmitted radar signals is to computationally increase the antenna aperture using the SAR algorithm used.

[0042] To minimize the computational effort required for this, the radar system in the vehicle can be switched from a first operating mode, in which a high-resolution radar image RB is obtained, to a second, computationally minimal operating mode. For example, such a high spatial resolution cannot be achieved in the second operating mode; however, in this second operating mode, environmental information can be acquired with less computational effort when calculating the radar information, which makes it possible to limit the detection range to a relevant area.

[0043] For example, the radar system can be converted from a first operating mode, which uses a SAR algorithm to calculate radar information, to a second operating mode in which radar information is calculated using a beamforming algorithm (e.g., digital beam forming algorithm, DBF). In other words, to acquire environmental information, an operating mode is selected in which the radar sensor influences the transmit and receive characteristics of the antenna by phase shifting the transmit and receive signals applied to the individual antennas of radar sensor 1, thus enabling it to capture the area surrounding radar sensor 1. This type of radar information acquisition requires significantly fewer computing resources.However, conclusions can be drawn from the radar information acquired in the second operating mode as to which areas contain relevant back reflections and which areas lack any. This allows the area for which a high-resolution radar image RB is calculated in the first operating mode to be limited to a relevant detection range, for example, by excluding regions where no reflections are present, which presumably represent open areas.

[0044] In other words, based on the radar information acquired in the second operating mode, areas are defined in which high-resolution radar information is to be calculated, as well as other areas in which the calculation of high-resolution radar information is not to be carried out for computational reasons. Based on the respective results of the environmental detection, either a contiguous area can be defined in which high-resolution radar information is to be calculated, or several non-contiguous sub-areas. In the radar image RB according to Fig. 2, the area for which high-resolution radar information is calculated is shown hatched. This area surrounds the objects O identified during the acquisition of environmental information.

[0045] Based on the definition of the relevant detection area(s), the calculation of the high-resolution radar information can be restricted accordingly. For example, the maximum possible detection area, which has M × N pixels, can be restricted by m pixels in a first spatial direction and n pixels in a second spatial direction, so that only one calculation of radar information for (Mm) × (Nn) pixels is necessary.

[0046] Furthermore, limiting the relevant detection area can result in a reduced number of individual measurements being required instead of L, for example, I fewer individual measurements. In other words, limiting the detection area can result in only (LI) individual measurements being required to calculate the high-resolution radar image RB. Thus, by defining the relevant detection area based on the acquired environmental information, the number of pixels calculated using (LI) individual measurements is advantageously reduced to (Mm) × (Nn) pixels.

[0047] To acquire environmental information, the radar system can be switched to the second, computationally minimal operating mode at specified intervals or as needed (e.g., depending on the vehicle's speed in space, etc.). This mode allows the radar system to acquire environmental information to define the relevant detection area. The radar system can then be operated again for a certain period of time in the first operating mode, in which high-resolution radar information is acquired in the detection area deemed relevant.

[0048] It is also possible to compare the environmental information obtained in the second operating mode of the radar system with existing additional environmental information stored in a memory unit, for example, in the form of map information, or to supplement the environmental information obtained by the radar system accordingly based on the additional environmental information. This advantageously reduces switching between the various operating modes and allows the environmental information to be provided with greater accuracy.

[0049] Fig. 3 shows a further embodiment of a method sequence for calculating a radar image based on a SAR algorithm, in which the computational effort is reduced by restricting the detection range based on environmental information, wherein already existing environmental information, for example information from a map, etc., or environmental information detected by at least one further sensor system is used.

[0050] Below, only the differences to the Fig. 2 described embodiment. Otherwise, the above mentioned instructions regarding the Fig. 2 also apply to the information provided in connection with Fig. 3 described procedures.

[0051] The main difference of the Fig. 3 schematically illustrated method to the previously described method according to Fig. 2 is that environmental information is used that was not detected by the radar sensor itself, but is either detected by a sensor system independent of the radar sensor or is based on environmental information available in the vehicle.

[0052] For example, an additional sensor system present in the vehicle can be used to capture environmental information around the vehicle and use this environmental information to define the relevant detection area. The additional sensor system can comprise a sensor that captures image information, for example, in the form of a camera (CCD sensor (CCD: charge-coupled device) or similar), or a sensor that scans the environment, such as a laser scanner or similar.

[0053] The additional sensor system can be operated continuously or intermittently in parallel with the radar system in order to be able to continuously or intermittently define the relevant detection area as the vehicle moves.

[0054] Alternatively, or in addition to using the additional sensor system to acquire environmental information, environmental information can be provided in the vehicle, for example, as map information stored in a memory unit or loaded into the vehicle via a data connection. This map information can be used to define the relevant detection area.

[0055] It is also possible to combine map information and environmental information obtained by other sensor systems in order to determine the relevant detection area based on both information sources.

[0056] After defining the relevant detection area, the high-resolution radar information can be calculated based on this relevant detection area. For example, by defining the relevant detection area based on the acquired environmental information, it is possible to calculate only high-resolution radar information for (Mm) × (Nn) pixels using (L1) individual measurements, where m and n represent the restriction of the number of pixels and L represents the reduction of the number of individual measurements.

[0057] Fig. 4 shows a schematic block diagram of a method for reducing computing time when calculating radar information in a moving vehicle.

[0058] First, environmental information in the local area detectable by the radar system is acquired or provided (S10). This environmental information preferably includes information that allows a statement to be made about the extent to which a spatial or surface section in the local area detectable by the radar system is likely to contain reflections of the radar signal that are relevant for a high-resolution radar image, for example, information relevant for roadway detection, objects with which a collision is to be avoided, etc.

[0059] Based on the acquired or provided environmental information, a detection area relevant for detection by the radar system is then determined (S11). For example, open-space detection can be performed, i.e., rooms or areas where no or only minimal reflections are expected are excluded as irrelevant areas from the detection area for which a high-resolution radar image is created.

[0060] After defining the detection range, radar information is calculated in a spatially restricted area based on the determined relevant detection range (S12). The computational effort can be reduced by spatially restricting the pixels for which the high-resolution radar information is calculated, as well as by reducing the number of individual measurements required to calculate high-resolution radar information.

[0061] The above assumption was made that the detection area is limited to a relevant detection area in two spatial directions (i.e., two-dimensionally) based on environmental information. It is understood that a spatial restriction (three-dimensionally) of the detection area is also possible based on the environmental information.

[0062] The invention has been described above using exemplary embodiments. It is understood that numerous changes and modifications are possible without departing from the scope of protection defined by the patent claims. List of reference symbols 1 radar sensor 10 vehicles A, A' Detection range of a single measurement O Object RB radar image

Claims

[1] Method for reducing computing time when calculating radar information in a moving vehicle, comprising the following steps: - detecting or providing environmental information in the local area (A, A') detectable by the radar system; - Determining a detection area relevant for detection by the radar system based on the environmental information detected or provided; - Calculating radar information in a locally restricted area based on the determined relevant detection range. The radar information is calculated in the radar system according to a synthetic aperture radar (SAR) algorithm, and the radar system is designed to acquire radar information based on sequential individual measurements. characterized bythat, based on the determined relevant detection range, at least some of the information from individual measurements taken at different locations is disregarded when calculating the radar information. [2] Method according to claim 1, characterized by that the environmental information is provided as map information or as map-like information. [3] Method according to claim 1 or 2, characterized by that the environmental information is at least partially recorded by at least one additional sensor system independent of the radar system. [4] Method according to claim 3, characterized by that the additional sensor system comprises a camera, a laser scanner or another sensor suitable for collecting environmental information. [5] Method according to one of the preceding claims, characterized bythat the environmental information is at least partially captured by the radar system itself. [6] Method according to claim 5, characterized by that the radar system for detecting environmental information is operated in a mode that minimizes computational effort. [7] Method according to claim 5 or 6, characterized by that the radar system is switched intermittently between a first operating mode in which high-resolution radar information is calculated, and a second operating mode in which radar information is calculated with a lower resolution compared to the first operating mode. [8] Method according to one of the preceding claims, characterized by that the environmental information collected or provided is evaluated with regard to local areas relevant for the collection of radar information. [9] Method according to claim 7, characterized bythat the areas relevant for the acquisition of radar information have at least one object (O) with an object size above an object size threshold. [10] Method according to claim 7, characterized by that the areas relevant for the acquisition of radar information are determined based on acquired radar information in the reduced computational effort operating mode that are above a threshold value. [11] Computer program product for reducing computing time in the calculation of radar information in a moving vehicle, wherein the computer program product comprises a computer-readable storage medium with program instructions, wherein the program instructions are executable by a processor to cause the processor to carry out a method according to one of the preceding claims. [12] Radar system for a vehicle comprising a radar system for calculating radar information in a moving vehicle, the radar system being designed to: - to receive environmental information relating to a local area detectable by the radar system; - to determine a detection area relevant for the radar system based on the environmental information received; and - to calculate radar information in a locally restricted area based on the determined relevant detection range; wherein the calculation of the radar information in the radar system is carried out according to a synthetic aperture radar algorithm (SAR algorithm), and wherein the radar system is designed to acquire radar information based on chronologically successive individual measurements, characterized bythat the radar system is designed to at least partially disregard the information from individual measurements carried out at different locations when calculating the radar information, based on the determined relevant detection range.

Citation Information

Patent Citations

  • Method for detecting object detection using radar sensor, involves discarding signals of irrelevant surrounding areas and supplying signals for high-resolution angle processing, to relevant surrounding areas

    DE102012020850A1

  • Environment model with adaptive grid

    DE102013018315A1

  • Providing an efficient environmental map for a vehicle

    DE102013207904A1

  • driver assistance system

    DE102015221911A1

  • Method for operating radar sensors in a motor vehicle and motor vehicle

    EP3151035A1