Modified ray tracker for electromagnetic response simulator

Abstract

This document describes techniques and systems for a modified ray tracker for an electromagnetic response simulator. Electromagnetic ray information is received, including a starting point and a direction. By converting the electromagnetic radiation information from a global coordinate system of the environment to a local coordinate system of the potential target, the potential target to be hit by the electromagnetic radiation can be determined. If it is calculated that the surface of the potential target is hit by the ray, the potential target is hit by the electromagnetic ray. The calculations performed in the local coordinate system of the potential target include a simplified large element physical optical formula of parallel rays. If it is determined that the face of the potential target is hit, an electromagnetic response associated with the potential target may be calculated. In this way, an efficient and accurate electromagnetic response model can be approximated.

Description

[0001] This application is a divisional application of the invention patent application with application number 202211197356.X and invention title "Modified Ray Tracker for Electromagnetic Response Simulator". Background Technology

[0002] Radar is a useful device for detecting and tracking objects. Therefore, radar offers numerous advantages for autonomous driving applications or driver assistance applications. During engineering and development, radar simulations can be run to evaluate performance under various conditions. These simulations can benefit from incorporating electromagnetic response models, which simulate the electromagnetic responses associated with radar in a simulated environment. However, existing electromagnetic response models may be too complex or too slow when used in some computer simulation environments; and if they must be used, some may not accurately simulate real-world electromagnetic responses. Summary of the Invention

[0003] This document describes the technology and system for a modified ray tracker used in an electromagnetic response simulator. Electromagnetic ray information, including the starting point and direction, is received. By transforming the electromagnetic ray information from the global coordinate system of the environment to the local coordinate system of the potential target, it can be determined that the potential target has been hit by an electromagnetic ray. If it is calculated that a surface of the potential target has been hit by the ray, then the potential target has been hit by an electromagnetic ray. The calculations performed in the local coordinate system of the potential target include simplified large-element physical optics formulas for parallel rays. If it is determined that a surface of the potential target has been hit, the electromagnetic response associated with the potential target can be calculated. In this way, an efficient and accurate electromagnetic response model can be approximated.

[0004] The aspects described below include a modified ray tracker for an electromagnetic response simulator. In one example, the method includes receiving first electromagnetic ray information of electromagnetic rays simulated in an environment comprising multiple potential targets. The first electromagnetic ray information includes the starting point and direction of the electromagnetic ray relative to a global coordinate system of the environment, and the corresponding ray density (ray beamwidth). The method further includes identifying the potential target among the multiple potential targets that is closest to the first electromagnetic ray information. The method further includes converting the first electromagnetic ray information into second electromagnetic ray information based on the potential target. The second electromagnetic ray information includes the starting point and direction of the second electromagnetic ray relative to a local coordinate system of the potential target. The method further includes determining whether the electromagnetic ray hits a surface of the potential target based on the second electromagnetic ray information and a pre-computed acceleration data structure indicating the geometric contour of the potential target. The method further includes determining whether to compute the electromagnetic response of the potential target based on whether the electromagnetic ray hits a surface of the potential target.

[0005] In another example, a system comprising at least one processor is described, the processor being configured to perform the methods summarized above and other methods described herein. This document also describes a non-transitory computer-readable storage medium having instructions that, when executed, cause the processor to perform the methods summarized above and other methods described herein.

[0006] This invention provides a simplified concept of a modified ray tracker for an electromagnetic response simulator, which is further described in the detailed description and accompanying drawings. This invention is not intended to identify essential features of the claimed subject matter, nor is it intended to define the scope of the claimed subject matter. Attached Figure Description

[0007] This document describes in detail one or more aspects of a modified ray tracker for an electromagnetic response simulator with reference to the following figures: Figure 1 An example environment according to the technology of this disclosure is shown, in which a radar simulator uses a modified ray tracker for electromagnetic response simulators to simulate one or more components of a radar system; Figure 2-1 An example electromagnetic simulator according to the technology of this disclosure is shown, which generates a modified ray tracker for an electromagnetic response simulator; Figure 2-2 An example sequence flowchart of an electromagnetic simulator according to the technology of this disclosure is shown, which outputs a modified ray tracker to a radar simulator for use in an electromagnetic response simulator. Figure 2-3 and Figure 2-4 The propagation path and ray density of an electromagnetic simulator according to the technology of this disclosure are considered using a modified ray tracker for an electromagnetic response simulator. Figure 3 The simulation process for generating a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown; Figure 4-1 and Figure 4-2 An example implementation of a modified shooting and bouncing ray method for an electromagnetic response simulator using a modified ray tracker is shown, according to the technology of this disclosure; Figures 5-1 to 5-4 An adaptive ray emission process for a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown; Figure 6-1A flowchart illustrating a modified ray tracing process for a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown. Figure 6-2 An example of a pre-calculated acceleration data structure for a target in an electromagnetic simulator according to the technology of this disclosure is shown, which generates a modified ray tracker for an electromagnetic response simulator. Figure 7 A flowchart illustrating a modified ray tracker for calculating an electromagnetic response simulator according to the techniques of this disclosure is shown. Figures 8-1 to 8-3 Example results of an electromagnetic simulator according to the technology of this disclosure are shown, which generates accurate and efficient electromagnetic responses for sensor simulators compared to other commercial simulators; Figure 9 An example method for a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown; Figure 10 Another example method of a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown; and Figure 11 Another example method of a modified ray tracker for an electromagnetic response simulator according to the technology of this disclosure is shown. Detailed Implementation

[0008] Overview

[0009] Automotive radar environment simulation tools use computers to simulate the wave propagation and scattering phenomena of millimeter-wave (MMW) radar signals in traffic scenarios, as well as the propagation of such signals through radar hardware and processing by the software backend. Compared to road testing, radar environment simulation is much less expensive and easier to repeat in certain key scenarios. Moreover, it provides deep insights into the interaction between the environment and radar signals, and therefore, it aids in and accelerates radar development. The propagation and scattering aspects of this simulation utilize electromagnetic simulation.

[0010] There are many existing electromagnetic simulation methods, which can be divided into three main categories: full-wave numerical methods, asymptotic methods, and model-based methods. Full-wave methods apply the integral or differential equations of Maxwell's equations, and are therefore very accurate. However, because traffic targets are much larger than the wavelength at the frequency of automotive radar, these methods are computationally prohibitive and extremely slow. If some acceleration techniques are applied, their time complexity becomes approximately O(n^2). 3 ) or O(n 2The model-based approach pre-calculates the electromagnetic responses from different targets and then uses post-processing to estimate the aggregate response from the dynamic simulation scenario. It typically ignores multipath effects and is the fastest method (e.g., O(n) time complexity), but sacrifices accuracy.

[0011] Asymptotic numerical methods assume that the propagation and scattering of electromagnetic waves are similar to those of light at MMV frequencies. Some examples of asymptotic methods include geometric optics (GO), physical optics (PO), and bouncing ray tracing (SBR) methods. The SBR method can be viewed as a combination of GO and PO. Wave propagation is described using the ray tracing method (GO approximation), and wave scattering is calculated using the PO approximation. It offers good simulation accuracy and efficiency (e.g., O(n...). A good balance is achieved between the time complexity of log(n) and the time complexity of log(n).

[0012] The SBR (Scattered Radar Cross Section) method is typically used to calculate the radar cross section (RCS) of targets placed at long distances. The traditional SBR method involves several steps. First, the target's geometry is discretized into facets, and an acceleration data structure is constructed to sort these facets. Next, uniformly distributed rays are emitted towards the target, and a ray-tracing algorithm is applied to find the ray propagation and the intersection points between the rays and the target. Finally, the scattered field at each ray and each ejection point can be calculated using PO (Polarization Point), and these are combined to further calculate the target's RCS.

[0013] However, the SBR method may have drawbacks. One drawback is that the SBR method alone cannot accurately calculate vertical multipath effects. Three key issues relate to this drawback. The first is that if the scattering direction toward the ground is not the specular reflection direction on the illuminated surface, the scattered electric field (E-field) reflected by the ground may not be captured in the calculation. Another issue is that ray density is related to simulation accuracy; rays scattering toward the ground or another large reflector and reflecting toward the target may have too sparse a ray density for accurate calculation. The third issue is that in some existing SBR methods, the road surface is treated as the target, and an excessive number of rays are emitted to cover the entire road, leading to a significant increase in computational costs in terms of resources and execution time.

[0014] Another drawback is the inability to accurately calculate horizontal multipath effects. One reason is similar to the vertical multipath case described above, where the density of reflected rays is too low (e.g., the rays are too sparse). Furthermore, weak responses from the target, such as reverse and reciprocal paths, are ignored. These weak responses are physically present and can be interpreted as ghosting targets by real radar. Including them in the electromagnetic model is crucial for maintaining high fidelity in simulations.

[0015] The third drawback is that in dynamic scene simulation, generating acceleration data structures for multiple surfaces of multiple targets for each frame leads to increased processing time and resources.

[0016] In contrast, the techniques and systems described in this document overcome these drawbacks. To overcome the inaccuracies caused by vertical multipath effects, mirror theory and ray tracing are combined with the SBR method. Mirror theory and ray tracing can account for vertical multipath effects while excluding the detection of waves that cannot be sensed (e.g., waves blocked by the physical properties of the target, waves that bounce off the ground but do not reflect back to the sensor).

[0017] The inaccuracy of horizontal multipath effects can be improved by using an adaptive ray emission process and examining additional backscattered and reciprocal paths that electromagnetic rays can follow. The adaptive ray emission process identifies distant targets, which typically receive a much lower density of spatial rays compared to closer targets. This process increases the ray density emitted towards distant targets and targets illuminated by multipath. The adaptive ray emission process can also identify rays that will not be reflected back to the sensor and are not analyzed. By examining the additional backscattered and reciprocal paths that electromagnetic rays may follow using the adaptive ray emission process, multipath effects that are typically excluded from analysis can be taken into account.

[0018] A modified ray tracing process can be used to improve the problem of inefficient dynamic scene simulation, where the acceleration data structure can be reused across different frames of the simulation and different targets with the same geometry in the scene. Furthermore, for rays with parallelogram footprints, a simplified large-element PO formula can be applied to reduce ray density while maintaining good accuracy.

[0019] As described in this paper, an electromagnetic simulator can use both a computer's central processing unit and image processing unit to generate electromagnetic responses from a traffic scene. Scene information (e.g., the position, orientation, and velocity of all targets) and basic transmitter / receiver parameters (position, orientation, field of view, and operating frequency) are taken as input. The output consists of a focused ray containing polarized electromagnetic responses from different parts of all detectable targets. A reference... Figure 3 The simulation process is described. Furthermore, this document generally describes the technologies and systems within the context of radar sensors and radar simulators. However, the described technologies and systems can be used to evaluate the electromagnetic response of any high-frequency application, such as microwave applications, WiFi, and cellular phone applications.

[0020] By using the techniques summarized above, both vertical and horizontal multipath effects can be included in the electromagnetic model. Furthermore, undetected electromagnetic rays can be excluded from analysis. Acceleration data structures can be reused without requiring recalculation for each frame of the simulation. In this way, accurate and efficient electromagnetic response models can be generated for use in sensor simulators.

[0021] Example Environment

[0022] Figure 1 An example environment 100 according to the technology of this disclosure is shown, in which a radar simulator 102 uses a modified ray tracker for electromagnetic response simulators to simulate one or more components of a radar system 104. In the depicted environment 100, the radar system 104 is mounted to or integrated within a vehicle 106. The radar system 104 is capable of detecting one or more targets 108 that are close to the vehicle 106 and associated with a signal-to-noise ratio greater than a detection threshold of the radar system 104. Although shown as a car, the vehicle 106 may represent other types of motorized vehicles (e.g., motorcycles, buses, tractors, semi-trailers, or construction equipment), other types of non-motorized vehicles (e.g., bicycles), other types of rail vehicles (e.g., trains or trams), water vehicles (e.g., boats or ships), aircraft (e.g., airplanes or helicopters), or spacecraft (e.g., satellites). Generally, the radar system 104 can be mounted to any type of mobile platform, including mobile machinery or robotic equipment.

[0023] In the depicted implementation, radar system 104 is mounted on top of vehicle 106. In other implementations, radar system 104 may be mounted on the lower side, front side, rear side, left side, or right side of vehicle 106. In some cases, vehicle 106 includes multiple radar systems 104, such as a first rear-mounted radar system 104 positioned near the left side of vehicle 104 and a second rear-mounted radar system 104 positioned near the right side of vehicle 104. Generally, the location of one or more radar systems 104 can be designed to provide a specific field of view containing a region of interest in which target 108 may be present. Example fields of view include 360-degree fields of view, one or more 180-degree fields of view, one or more 90-degree fields of view, etc., which may overlap (e.g., four 120-degree fields of view).

[0024] Target 108 is composed of one or more types of material that reflect radar signals. Depending on the application, target 108 may represent a target of interest or clutter. In some cases, target 108 is a moving target, such as another vehicle, person, or animal. In other cases, target 108 is a stationary target, such as continuous or discontinuous road obstacles (e.g., traffic cones, concrete barriers, guardrails, or fences), trees, or parked vehicles.

[0025] Vehicle 106 can implement one or more radar-based systems utilizing data provided by radar system 104. Examples of radar-based systems include driver assistance systems or autonomous driving systems. A driver assistance system can use data provided by radar system 104 to monitor one or more blind spots of vehicle 106 and issue a warning to the driver of a potential collision with target 108. An autonomous driving system can move vehicle 106 to a location while analyzing data provided by radar system 104 to avoid collisions with other targets 108, perform emergency braking, change lanes, or adjust the speed of vehicle 106.

[0026] Radar system 104 can be implemented as a continuous wave or pulse radar, a frequency-modulated or phase-modulated radar, a single-input single-output (SISO) or multiple-input multiple-output (MIMO) radar, or some combination thereof. In the depicted configuration, radar system 104 is implemented as a frequency-modulated continuous wave (FMCW) MIMO radar employing a code division multiple access (CDMA) scheme. To detect target 108, radar system 104 transmits at least one radar transmit signal 110. At least a portion of the radar transmit signal 110 is reflected by target 108. This reflected portion represents radar receive signal 112. Radar system 104 receives and processes radar receive signal 112 to provide data to the radar-based system of vehicle 106.

[0027] To transmit radar transmit signal 110 and receive radar receive signal 112, radar system 104 includes at least one MIMO antenna 114 and at least one monolithic microwave integrated circuit (MMIC) 116. MMIC 116 includes circuitry and logic for transmitting and receiving radar signals via MIMO antenna 114. Components of MMIC 116 may include amplifiers, mixers, switches, analog-to-digital converters, or filters for conditioning the radar signals. MMIC 116 also includes logic for performing in-phase / quadrature (I / Q) operations, such as modulation or demodulation. MMIC 116 may include at least one transmitter and at least one receiver.

[0028] Some implementations of radar system 104 can use MIMO antenna 114 to form steered or non-steered, wide or narrow beams. Steering and shaping can be achieved through analog beamforming or digital beamforming. One or more transmit subarrays can have, for example, a non-steered omnidirectional radiation mode, or can generate wide, steered beams to illuminate a large spatial volume. To achieve target angular accuracy and angular resolution, receiver subarrays can include multiple receive antenna elements to generate hundreds of narrow steered beams using digital beamforming. In this way, radar system 104 can effectively monitor the external environment and detect one or more targets 108 within the region of interest.

[0029] The radar system 104 also includes one or more processors 118 and a computer-readable storage medium (CRM) 120. The CRM 120 includes radar software 122 for analyzing radar received signals 112, detecting targets 108, and determining one or more characteristics of the target 108 (e.g., position or velocity).

[0030] Radar simulator 102 simulates the hardware and / or software of radar system 104 to enable the performance of radar system 104 to be evaluated against a given simulated environment. For example, throughout the development cycle of radar system 104, radar simulator 102 can be used to evaluate different system designs (e.g., different hardware configurations or different operating modes), test different versions of radar software 122, or verify requirements. Using radar simulator 102 allows design or implementation issues within radar system 104 to be rapidly identified during the design, integration, and testing phases before field testing is performed.

[0031] Radar simulator 102 performs the operation of hardware (e.g., components of MIMO antenna 114 and / or MMIC 116), radar software 122 executed by processor 118, or a combination thereof, for simulating a radar system. Radar simulator 102 can also take into account non-ideal characteristics of radar system 104 or environment 100, such as noise or nonlinearity. Specifically, radar simulator 102 can model phase noise, waveform nonlinearity, and uncorrelated noise within MMIC 116. With these capabilities, radar simulator 102 can have a noise floor and dynamic range similar to radar system 104.

[0032] Radar simulator 102 includes at least one processor 124 and a computer-readable storage medium 126. The CRM 126 may be one or more memory devices capable of storing persistent and / or non-transient data. In some cases, the processor 124 and CRM 126 are packaged together within an integrated circuit or on a printed circuit board. In other cases, the processor 124 and CRM 126 may be implemented separately and operatively coupled together, such that the processor 124 can access instructions stored in the CRM 126.

[0033] CRM 126 includes at least one interface module 128 and at least one radar hardware simulator 130. Optionally, CRM 126 may also include at least one radar software simulator 132. The interface module 128, radar hardware simulator 130, and radar software simulator 132 can be implemented using hardware, software, firmware, or a combination thereof. In this example, processor 124 implements the interface module 128, radar hardware simulator 130, and radar software simulator 132.

[0034] Interface module 128 enables radar simulator 102 to process environmental response data provided by the electromagnetic simulator (as shown in Figure 2). Generally, interface module 128 transforms the environmental response data into a form usable by radar hardware simulator 130. In some cases, interface module 128 adjusts the environmental response data to account for the antenna response of MIMO antenna 114.

[0035] The radar hardware simulator 130 simulates the MMIC 116. Specifically, the radar hardware simulator 130 performs operations simulating waveform generation, modulation, demodulation, multiplexing, amplification, frequency conversion, filtering, and / or analog-to-digital conversion performed by the MMIC 116. In other words, the radar hardware simulator 130 executes the operations of the MMIC 116, which occur between the processor 118 and the MIMO antenna 114. The radar hardware simulator 130 can account for the dynamic range of the radar system 104, as well as the presence of nonlinear effects and noise. The radar hardware simulator 130 can be used to verify different hardware configurations and operating modes of the radar system 104.

[0036] Radar software simulator 132 can model the radar software 122 of radar system 104. Specifically, radar software simulator 132 performs digital baseband processing operations that simulate operations performed by processor 118. These operations may include Fourier transform (e.g., fast Fourier transform), noise basis estimation, clutter map generation, constant false alarm rate thresholding, object detection, and object location estimation (e.g., digital beamforming). In some cases, radar software simulator 132 includes a version of radar software 122. In this way, radar simulator 102 can be used to verify software requirements and evaluate different software versions of radar system 104.

[0037] During operation, radar simulator 102 receives electromagnetic response data from electromagnetic simulator 134 and generates a radar report. This radar report can be used to evaluate the performance of radar system 104 for a given simulated environment. By providing accurate electromagnetic response data, electromagnetic simulator 134 enables reliable evaluation of the performance of radar system 104.

[0038] The electromagnetic simulator 134 includes at least one processor 136 and a computer-readable storage medium 138. The CRM 138 may be implemented by one or more memory devices capable of implementing persistent and / or non-transient data storage. In some cases, the processor 136 and the CRM 138 are packaged together within an integrated circuit or on a printed circuit board. In other cases, the processor 136 and the CRM 138 may be implemented separately but operatively coupled together, such that the processor 136 can access instructions stored in the CRM 138.

[0039] CRM 138 includes a ray emitter 140 and a ray tracker 142. CRM 138 also includes an electromagnetic (EM) module 144. The ray emitter 140, ray tracker 142, and EM module 144 can be implemented using hardware, software, firmware, or a combination thereof. In this example, processor 136 implements the ray emitter 140, ray tracker 142, and EM module 144.

[0040] Ray emitter 140 enables electromagnetic simulator 134 to adaptively emit electromagnetic rays in the simulation. Ray tracker 142 enables electromagnetic simulator to track rays, including reflections from rays originating from targets in the simulation. EM module 144 calculates the electromagnetic response of each tracked ray. Ray emitter 140, ray tracker 142, and EM module 144 will be discussed in more detail below.

[0041] Figure 2-1An example electromagnetic simulator 134 is shown, which generates a modified ray tracer for use in an electromagnetic response simulator. The electromagnetic simulator 134 evaluates the propagation and scattering effects of simulated electromagnetic signals, such as radio frequency signals. Generally, the electromagnetic simulator 134 evaluates one or more propagation paths 204 that originate from and return to a reference point 206 within the simulation environment 208. Example types of propagation paths 204 may include a direct line-of-sight path 204-1 or an indirect path 204-2 associated with multipath propagation. In some cases, the electromagnetic simulator 134 uses ray tracing to determine the various propagation paths 204. (About...) Figure 2-3 and Figure 2-4 The multipath propagation and associated paths are further described.

[0042] During operation, the electromagnetic simulator 134 determines the characteristics of different propagation paths 204 within the simulation environment 208. Specifically, the electromagnetic simulator 134 receives scene information 210, which specifies the characteristics of one or more simulated targets (e.g., the position, orientation, velocity, and / or material composition of the simulated targets). The electromagnetic simulator 134 generates the simulation environment 208 based on the scene information 210.

[0043] As an example, simulation environment 208 includes simulated versions of vehicles 212, people 214, trees 216, and buildings 218. Scene information 210 also enables electromagnetic simulator 134 to simulate different types of weather 220, terrain, road geometry, traffic rules, and traffic conditions. In some cases, scene information 210 specifies the movement of reference point 206. In simulation environment 208, reference point 206 represents the position on simulated vehicle 222, which can be compared with... Figure 1 The location corresponds to the radar system 104 on the vehicle 106.

[0044] Electromagnetic simulator 134 generates environmental response data 224, which can change over time as a target within reference point 206 or the simulated environment 208 moves. Electromagnetic response data 224 can include composite data with amplitude, phase, and frequency information. Electromagnetic simulator 134 can rely on information about radar system 104 to generate environmental response data 224. Examples of information about radar system 104 may include the radar system 104's location, orientation, field of view, and operating frequency.

[0045] As an example, the electromagnetic simulator 134 can emit electromagnetic rays in several directions and use ray tracing to generate electromagnetic response data 224. To improve efficiency, rays can be emitted and received from a reference point 206, which may represent the center of the radar system 104. Propagation paths, including multipath effects, can be examined. The electromagnetic simulator 134 can exclude propagation paths that do not reflect back to the reference point 206 when generating the electromagnetic response data 224. By including multipath effects and excluding paths that do not reflect back to the reference point 206, the electromagnetic simulator 134 can accurately model the electromagnetic energy that the radar system can detect in the real world. The electromagnetic simulator 134 also uses an acceleration data structure based on a computer-aided design (CAD) model of the simulated target (e.g., indicating geometric contours). The acceleration data structure includes information about the surface of the target. Targets using the same CAD model can use the same pre-calculated acceleration data structure. For example, multiple vehicles 212 can use the same pre-calculated acceleration data structure. Furthermore, if the target's geometry is static, the pre-calculated acceleration data structure is not recalculated during the duration of the simulation. In other words, the pre-computed acceleration data structure can be used for a portion of the simulation or for the entire simulation (e.g., for each frame of the simulation). The acceleration data structure for a target with dynamic geometry (e.g., human 214) is recomputed for each frame. This allows for more efficient use of computational resources.

[0046] Electromagnetic simulator 134 can provide propagation data 226 as electromagnetic response data 224. Propagation data 226 includes a list of propagation paths 204 (e.g., rays) and characteristics of the electromagnetic signals propagating along the propagation paths 204. These characteristics may include relative amplitude, Doppler shift, time of flight, or departure and angle of arrival. In this example, electromagnetic simulator 134 can generate propagation data 226 without relying on information about radar system 104.

[0047] Electromagnetic simulator 134 can provide integrated data 228 as electromagnetic response data 224. Instead of including information about individual propagation paths 204, integrated data 228 represents the combined environmental response generated by multiple propagation paths 204 of at least one transmit-receive channel pair. To generate integrated data 228, electromagnetic simulator 134 receives radar parameters 230 during operation. Radar parameters 230 provide information about radar system 104 or radar transmitted signal 110. Default radar parameters can specify ideal antenna response or ideal waveform characteristics. Electromagnetic simulator 134 transmits electromagnetic response data 224 to radar simulator 102 (… Figure 1 (of), such as about Figure 2-2 Further description.

[0048] Figure 2-2 An example sequence flowchart 200 of an electromagnetic simulator 134 is shown, which outputs a modified ray tracker for an electromagnetic response simulator to a radar simulator. The electromagnetic simulator 134 simulates environment 208 based on scene information 210. Using the techniques described herein, the electromagnetic simulator 134 can output highly accurate electromagnetic response data 224 to the hardware simulator 130 and / or software simulator 132 of the radar simulator 102. During operation, the radar simulator 102 analyzes the electromagnetic response data 224 and generates a radar report 232, which can be used to evaluate the performance of the radar system 104 in the simulated environment 208. Different types of radar reports 232 can be possible. Generally, the radar report 232 represents any type of data that can be generated by the radar system 104. This can include signal samples received by the MMIC 116, information processed by the processor 118, or information from radar-based systems transmitted to the vehicle 106.

[0049] Figure 2-3 and Figure 2-4 The diagram illustrates the propagation path and ray density considered by the electromagnetic simulator 134 using a modified ray tracker technique for electromagnetic response simulators. (As shown...) Figure 2-3 As shown in (a), the electromagnetic simulator 134 considers the propagation path (e.g., scattered E field) 234 originating from the transceiver 236 and reflected from the target 238 and the ground plane 240. Figure 2-3 (b) shows a sparse ray density 242 on target 238, which may be too sparse to accurately model electromagnetic reflection in the real world. In this case, electromagnetic simulator 134 can detect the defect and emit rays with a higher ray density toward target 238.

[0050] Electromagnetic simulator 134 considers five propagation paths with multipath effects, such as... Figure 2-4 As shown. Figure 2-4 (a) shows path 224, which directly reflects back to transmitter 236 if the path is not blocked by other surfaces. Figure 2-4 (b) shows path 246 of the reflective transceiver mirror 248. This corresponds to the path from ground to transceiver 236. Figure 2-4 In (c), path 250 is reversed on the incident path. If path 250 is not the mirror reflection direction of the target, then path 250 is considered. Figure 2-4 (d) and Figure 2-4 (e) are respectively Figure 2-4 (a) and Figure 2-4(b) shows the inverse routes of paths 244 and 246. That is, paths 252 and 254 have the same electromagnetic response as paths 244 and 246, but are in opposite departure and arrival directions (e.g., reverse paths). By evaluating these five additional paths, the electromagnetic simulator 134 can more accurately simulate the real world.

[0051] Example Implementation

[0052] Figure 3 The simulation process 300 for generating a modified ray tracker for an electromagnetic response simulator is illustrated. At step 302, the simulation scene and transmitter / receiver information are generated by the ray emitter (e.g., Figure 1 The scene information (140) is read from the ray emitter. Scene information may include the position, orientation, and velocity of all targets to be simulated. The geometric profile of each target may also be included in the scene information. Emitter / receiver information may include some basic parameters, including position, orientation, field of view, and operating frequency.

[0053] At step 304, the ray emitter emits electromagnetic rays with varying densities toward the target and its mirror image. The ray density in a particular direction is determined by the range of the target the electromagnetic ray might hit. The mirror image of the target is the reflection of the target about the ground plane. In the simulation, the ground plane can be the ground surface or some other barrier reflecting the electromagnetic rays in the simulation (e.g., buildings, fences, any surface that contributes to multipath reflection). One method for creating and analyzing the mirror image of the sensor, the target, and the scattered E-field utilizes electromagnetic mirror theory. Figures 5-1 to 5-4 The adaptive ray emission process is further described.

[0054] At step 306, the ray tracker (e.g., Figure 1 The ray tracker 142 in the process performs ray tracing. This step involves a modified ray tracing procedure to calculate the locations of ray impact and reflection. (Regarding...) Figure 6-1 and Figure 6-2 The modified ray tracing process is further described.

[0055] At step 308, the EM module (e.g., Figure 1 The EM module (144) performs electromagnetic calculations. The EM module calculates the polarization electromagnetic response of each ray. (About...) Figure 7 The process of the EM module is further described.

[0056] In step 310, electromagnetic rays with similar properties are combined into a focused ray. Then, in step 312, the focused ray is output to the sensor simulator.

[0057] Figure 4-1 and Figure 4-2An example implementation of a modified SBR method using a modified ray tracker for an electromagnetic response simulator is shown. The modified SBR method determines the electromagnetic rays interacting with target 402. Electromagnetic rays can have a forward propagation path to the target and a backward propagation path from the target. Electromagnetic rays are determined by considering electromagnetic rays emitted towards the target and reflected directly back to the receiver, as well as electromagnetic rays including multipath effects. The method uses the GO process to approximate the multipath effects produced by electromagnetic rays reflected from the ground. In some implementations, electromagnetic mirror theory is employed to determine multipath effects originating from the ground (e.g., the ground plane) and enables the electromagnetic simulator (e.g., Figure 1 The electromagnetic simulator 134 in the text can include multipath electromagnetic rays, which would otherwise be ignored.

[0058] Using electromagnetic mirror theory, the process involves emitting initial electromagnetic rays from the radar toward the target and its mirror image at the same spatial density. A ray-tracing process is applied to the electromagnetic rays, where, in addition to the direct propagation path without multipath effects, the electromagnetic rays bouncing between the target and their respective mirror images are considered. The scattered E-field is evaluated at the location of the radar and its mirror image based on the equivalent surface current. Some paths of the incident and scattered fields may cross the ground plane. The bulk dielectric Fresnel reflection coefficient is calculated and multiplied by the scattered field and the corresponding incident field.

[0059] like Figure 4-1 As shown, target image 406 of target 402 is mirrored with respect to ground plane 404. Similarly, radar sensor 408 has a corresponding radar sensor image 410 mirrored with respect to ground plane 404. Once target image 406 and radar sensor image 410 are determined, ground plane 404 can be removed for ray tracing calculations. Ray tracing can be performed without ground plane 404, including both target image 406 and radar sensor 410.

[0060] Electromagnetic ray 412 is emitted from radar sensor 408 towards target 402 and target mirror 406 with the same spatial ray density. Electromagnetic ray 412-1 hits target 402 without multipath propagation and has a reflected electromagnetic ray 414-1 that returns directly to radar sensor 408. Electromagnetic ray 412-1 can also have a reflected electromagnetic ray 414-2 directed towards radar sensor mirror 410. However, the reflected electromagnetic ray 414-2 is blocked by target 402 and will not reach radar sensor mirror 410. Therefore, the reflected electromagnetic ray 414-2 can be discarded.

[0061] Electromagnetic ray 412-2 is emitted towards ground plane 404. By ignoring ground plane 404, electromagnetic ray 412-2 can be examined relative to target mirror image 406 and radar sensor mirror image 410. The dashed segment of electromagnetic ray 412-2 represents the segment that bounces off ground plane 404 and strikes target 402 below the rear bumper. Electromagnetic ray 412-2 is reflected directly from target mirror image 406 to radar sensor mirror image 410 as reflected electromagnetic ray 414-3. Electromagnetic ray 414-3 (a direct path without bounce after being reflected by target mirror image 406) is equivalent to reflected electromagnetic ray 414-4 (a path including one bounce before being reflected by target 402). Furthermore, reflected electromagnetic ray 414-5 is reflected from target 402 and ground plane 404 (bounces once after being reflected by target 402). The reflected electromagnetic ray 414-5 can be examined as an equivalent reflected electromagnetic ray 414-6 that propagates directly from the target 402 to the radar sensor mirror 410 (without bouncing after being reflected by the target 402).

[0062] Figure 4-2 Electromagnetic rays 416-1, including several bounces, are shown. Similar to... Figure 4-1 The electromagnetic ray shown simplifies the calculation of its propagation path by examining electromagnetic ray 416-1 relative to target mirror 406 and radar sensor mirror 410. Reflected electromagnetic ray 414-2, which bounces off ground plane 404, can be examined as an equivalent bounced electromagnetic ray 416-3 excluding the bounce. Furthermore, reflected electromagnetic ray 416-4 can be examined relative to target mirror 406, and its path includes one bounce from target mirror 406. The modified SBR method using electromagnetic mirror theory simplifies the calculations involving the scattered E-field and excludes any blocked electromagnetic rays.

[0063] In relation to Figure 4-1 and Figure 4-2In a non-limiting example, to determine the electromagnetic ray interacting with target 402, a target mirror image 406, mirrored with respect to ground plane 404 and target 402, is determined. Similarly, an electromagnetic sensor mirror image 410, mirrored with electromagnetic sensor 408, is determined. Electromagnetic ray 412 is emitted from electromagnetic sensor 408 (e.g., emitted ray) and has the same spatial ray density relative to target 402 and target mirror image 406. The dielectric Fresnel reflection coefficients of the target and ground plane are calculated based on electromagnetic ray 412. The scattered field and the corresponding incident field are then multiplied by the dielectric Fresnel reflection coefficients. The scattered field is evaluated by calculating the incident field of electromagnetic ray 412 on the target using geometrical optics. The scattered field and equivalent surface current of the reflected electromagnetic ray are calculated using physical optics. Any electromagnetic energy blocked by the physical properties of the target (e.g., electromagnetic ray 414-2) is excluded. This allows for the evaluation of only multipath reflections associated with target 402, without evaluating blocked electromagnetic energy that is undetectable by electromagnetic sensor 408.

[0064] Figures 5-1 to 5-4 An adaptive ray emission process 500-1 for a modified ray tracker used in an electromagnetic response simulator is illustrated. During ray emission, the direction and density of the electromagnetic rays (e.g., in the azimuth plane) are determined. This is crucial for both accuracy and simulation speed, as they are directly related to the complexity of the analysis. In many conventional SBR implementations, targets are illuminated by plane waves (which are typically far from the source / receiver), and the ray density of the electromagnetic rays is usually kept constant to produce consistent accuracy across targets. The ray emission process used in conventional SBR methods cannot be directly applied to millimeter-wave automotive radar applications, where radars with wide field of view are not necessarily at long range from targets present in traffic scenarios. Furthermore, the accuracy of the simulation is highly dependent on the desired spatial ray density on the target surface. Therefore, rays hitting near and far targets must have different angular ray densities to maintain the same simulation accuracy. Without considering the geometric complexity of the target, for constant accuracy, the angular ray density... It should be a function of the ray length d:

[0065] In adaptive ray emission, the length is accounted for multiple bounces if multipath propagation is considered. Multipath effects from roads, large vehicles, or guardrails are common and significant in automotive radar applications, and these are addressed in adaptive ray emission. Ground multipath is a special case of multipath because the ground is typically flat and acts like a mirror in the GO approximation, exhibiting some reflection loss to the electric field. For scenarios with a ground plane, emitting all portions of the electromagnetic ray to the ground is computationally inefficient. Instead, electromagnetic rays are emitted to the ground only if the reflected electromagnetic ray can hit the target. The angular ray density of the electromagnetic ray emitted to the ground can be determined by the length of the electromagnetic ray after reflection from the ground to the point of impact on the target, rather than the length of the electromagnetic ray to the point of impact on the ground, since the target is the object of interest in the simulation. Other multipath effects (such as horizontal multipath from large vehicles or guardrails) cannot be explicitly given, but the two-stage ray emission process 500-1 can accurately simulate these effects.

[0066] Figure 5-1 A flowchart of the adaptive ray emission process 500-1 is shown. At step 502, target information and the field of view of the emitter / receiver are received for the ray emitter to simulate. The target information includes the geometric information of all targets in the environment.

[0067] At step 504, the corner occupancy grid of the target is determined, and sparse electromagnetic rays are emitted towards the occupied pixels of the corner occupancy grid. To determine the corner occupancy grid, the three-dimensional boundaries of each target and each corresponding target mirror are determined, and the target's occupancy grid is calculated with respect to the transmitter / receiver. In one example, the resolution of the corner occupancy grid is 0.1 degrees. The sparse electromagnetic rays are uniformly distributed as they are directed towards the center of each pixel in the corner occupancy grid occupied by the target and its mirror (e.g., based on mirror theory).

[0068] At step 506, ray tracing is performed by the ray tracker, and the distance the sparse electromagnetic rays travel to reach the corresponding target up to the second bounce of each electromagnetic ray (e.g., electromagnetic energy). Up to the second bounce means that this distance does not include sparse rays with more than one multipath reflection. The ray tracing process is applied to calculate the length of each electromagnetic ray up to the second bounce (e.g., the distance the electromagnetic ray travels from the source to the second bounce point if a second bounce point exists). That is, the length of each electromagnetic ray can include a first distance from the electromagnetic sensor to the first reflection point and a second distance from the first reflection point to the second reflection point.

[0069] At step 508, the angular ray density of each pixel is determined based on the length of the sparse electromagnetic rays emitted at step 504. By determining the angular ray density of each pixel, targets with lower angular ray densities and their corresponding mirror images can be identified. The angular ray density is calculated by dividing the desired spatial ray density by the length of the sparse electromagnetic rays.

[0070] At step 510, dense electromagnetic rays are emitted. Taking multipath effects into account, dense electromagnetic rays are emitted with different angular ray densities for different targets. The adaptive ray emission process 500-1 also considers... Figure 2-4 The propagation path is described. The adaptive ray emission process 500-1 enables electromagnetic simulations to have higher fidelity, resulting in higher accuracy.

[0071] Figure 5-2 It shows about Figure 5-1 Example implementation 500-2 of step 504. Transceiver 512 uses a ray emitter to emit sparse electromagnetic rays with equal angular ray density toward target 514. For example... Figure 5-2 As shown, more sparse electromagnetic rays hit the closer targets 514-1 and 514-3, rather than the more distant target 514-2. Furthermore, considering the mirror image of target 514 (not shown), some sparse electromagnetic rays are emitted towards boundary 516. (As shown...) Figure 5-2 As shown, electromagnetic ray 518 did not hit the target. Electromagnetic ray 518 represents an electromagnetic ray that travels beyond the range of transceiver 512 and can be ignored.

[0072] Figure 5-3 It shows about Figure 5-1 Step 506 is an example implementation of 500-3. Several multipaths of sparse electromagnetic rays are shown. The length of the sparse electromagnetic ray is calculated up to the second bounce. For example, electromagnetic ray 520 has two components. The total length of electromagnetic ray 520 is found by adding the lengths of these two components.

[0073] Figure 5-4 It shows about Figure 5-1 Example implementation 500-4 of step 508. Calculate the angular ray density of each of electromagnetic rays 520, 522, and 524 based on a predefined spatial ray density (e.g., 0.01 m). If the two components of electromagnetic ray 520 have lengths r1 and r2, and the predefined ray density is expressed as d... req ,but

[0074] Where ρ = angular ray density.

[0075] When electromagnetic ray 524 has only one component, the angular ray density can be calculated as follows:

[0076] Where r1 is the length of one component of electromagnetic ray 524.

[0077] The adaptive ray emission process 500-1 assumes that distant targets (e.g., 514-2) may have a lower angular ray density than closer targets (e.g., targets 514-1 and 514-3). The adaptive ray emission process 500-1 calculates the angular ray density and emits dense electromagnetic rays based on the angular ray density calculated using the length of sparse electromagnetic rays. In this way, the adaptive ray emission process 500-1 enables distant targets to have similar resolution to closer targets.

[0078] Figure 6-1 A flowchart of a modified ray-tracking process 600-1 for a modified ray tracker used in an electromagnetic response simulator is shown. At step 602, the ray tracker receives information about the emitted electromagnetic ray. This information includes the starting point and direction of travel of the electromagnetic ray relative to the global coordinate system of the simulation environment.

[0079] At step 604, the ray tracker searches for the nearest target that the electromagnetic ray might hit. The search is performed based on global coordinates of the simulated environment. In one example, the search is performed between targets in a two-dimensional KD tree. At step 604, a target is considered to be a rectangle that defines the target.

[0080] At step 606, once the ray tracker detects a potential target that may be struck by the electromagnetic ray, it converts the information of the electromagnetic ray into the target's local coordinates. In one example, information about the target is stored in an acceleration data structure that indicates the target's geometric profile. The acceleration data structure can be pre-computed and used for any target with the same geometric profile. Furthermore, the pre-computed acceleration data structure can be used throughout the duration of the simulation. The duration can include the entire simulation or a portion of it.

[0081] At step 608, the ray tracker searches for the nearest surface of the target it might hit. Surface information can be stored in an acceleration data structure. The ray tracker compares the information of electromagnetic rays in the target's local coordinates with the target's surface information.

[0082] At step 610, if the ray tracker determines that the target's face was hit by the electromagnetic ray after comparing the face information with the electromagnetic ray, the ray tracker proceeds to step 612. If the electromagnetic ray did not hit the target's face, the ray tracker proceeds to step 614.

[0083] At step 612, the information of the reflected electromagnetic ray is converted to simulate the global coordinates of the environment. The reflected ray information includes the point of impact of the electromagnetic ray on the target's surface (e.g., the origin of the reflected electromagnetic ray) and the direction of reflection. The reflected ray information is output to the EM module. Following the step description, the process of determining whether the target's surface was hit by an electromagnetic ray and the direction of the reflected electromagnetic ray is described in more detail.

[0084] At step 614, the ray tracker determines whether all potential targets have been tested. If no targets remain to be tested, the ray tracker proceeds to step 616. If more potential targets are to be tested, the ray tracker proceeds to step 618.

[0085] At step 616, if the electromagnetic ray does not hit any target, the ray tracker outputs the condition that the electromagnetic ray did not hit the target. This electromagnetic ray can be ignored.

[0086] At step 618, ignoring the potential target being evaluated, the ray tracker begins the process at step 604. These steps ensure that the output of the electromagnetic simulator includes only reflected electromagnetic rays.

[0087] The process of determining whether a target surface has been struck by electromagnetic rays depends on the following principle. Wave propagation at millimeter-wave frequencies is approximated by GO (Glass Orb). The principle is relatively simple: rays travel in straight lines, and if they strike a reflective surface, they are reflected and propagate along the mirror direction.

[0088] in It is the direction of incidence. It is the normal vector on the surface. Indicates the direction of reflection.

[0089] In SBR simulations, after the ray tracing process, each surface region hit by the ray tube generates an equivalent surface current based on boundary conditions. This equivalent surface current radiates the scattered field back to the receiver. Depending on the scenario, there may be many paths for the scattered field to reach the receiver.

[0090] According to Huygens' principle, the scattered field from the current source is given by the following equation:

[0091] in and Represents current and magnetic surface current. It is the wavenumber in free space. It is an idempotent factor of the parallel vector. From the source To the observation point Vector length:

[0092] If the distance R is much greater than the wavelength (i.e., Then, the long-distance approximation can be applied to (1):

[0093] Surface current and Defined as:

[0094] in and These are the incident electric field and magnetic field, respectively. It is the tangential vector on the surface, defined as , It is the unit vector of the incident direction. It is the surface normal vector. and It is the Fresnel reflection coefficient of the surface under transverse magnetic (TM) and transverse electric (TE) conditions.

[0095] Fresnel reflectance and Given by the following equation,

[0096] in and It is the characteristic impedance of air and the target. and These are the angles of incidence and angles of refraction relative to the target surface.

[0097] For metallic materials Very large Become negligible and . ,and Equation (7) can be further simplified to:

[0098] If the radar is in a polygonal area At longer distances, large element PO can be applied. The surface current can then be rewritten as:

[0099] in This indicates the location of the emitter. Then, the scattered field becomes:

[0100] in , yes Components in direction, , It is the component on the surface of the surface. . This indicates the number of sides of the polygon. It is perpendicular to The tangential unit vector on the surface. Representing the edge vectors of a polygon:

[0101] In the SBR ray approximation, the ray tube has a square shape, and the integration region on the target surface is the projection of the ray, assumed to be a parallelogram.

[0102] For a parallelogram, the summation term in (15) can be simplified to:

[0103] This allows for the efficient calculation of the scattered field of the current induced by the ray tube in scattering directions with different polarizations.

[0104] Figure 6-2 An example of a pre-computed acceleration data structure 620 for a target in an electromagnetic simulator is shown, which generates a modified ray tracker for an electromagnetic response simulator. The geometry of many targets in the electromagnetic simulation remains unchanged throughout the simulation. Only information related to position or orientation may change. This allows these targets to use the pre-computed acceleration data structure. For these targets, the pre-computed acceleration data structure 620 can be used in every frame of the simulation. Furthermore, any target with the same CAD model can use the pre-computed data structure 620. Local coordinates of the target's CAD model 622 are used to generate the pre-computed acceleration data structure 620. The target's material properties can be loaded to generate a polarization reflection coefficient lookup table, which can be used in subsequent steps of the electromagnetic calculations. Similarly, if the material is penetrable at a given thickness (represented by a threshold), the target's transmission coefficient can be calculated. The pre-computed acceleration data structure 620 makes the simulation efficient and saves computational resources such as processing and memory.

[0105] Figure 7 A flowchart 700 is shown for calculating a modified ray tracker used in an electromagnetic response simulator. At step 702, the EM module receives geometric information about the electromagnetic rays after ray tracing is performed. The EM module can receive information from the ray tracker.

[0106] At step 704, the EM module calculates the equivalent surface current J for each bounce of the electromagnetic ray. m and M m J m and M m This includes both V-polarization and H-polarization cases. The surface current is based on the incident field with different polarizations.

[0107] At step 706, the EM module examines up to five scattering paths returning to the transceiver (e.g., scattering paths 244, 246, 250, 252, and 254 in Figures 2-4). If a scattering path is required, the EM module calculates the scattering field for each required path with different polarizations. At step 706-1, the EM module calculates paths such as paths 244 and 252. At step 706-2, the EM module calculates paths such as paths 246 and 254. At step 706-3, the EM module calculates paths such as path 250.

[0108] At step 708, the EM module records the scattered E-field of each electromagnetic ray and their departure and arrival directions. The EM module can then output the records to the radar simulator.

[0109] Figures 8-1 to 8-3 Example results from the electromagnetic simulator are shown, demonstrating its ability to generate accurate and efficient electromagnetic responses for sensor simulators compared to other commercial simulators. To demonstrate simulation accuracy considering vertical multipath, a radar cross-section of a corner reflector placed approximately 0.52 meters above a road (e.g., the thin line in each figure) is simulated and compared to measurements from a real radar (e.g., the thick line in each figure). All simulations use the same high-fidelity radar simulator. Figure 8-1 and Figure 8-2 This is a graph showing different example results of commercially available electromagnetic simulators. Figure 8-3 This is a graph showing the results of using an electromagnetic simulator based on the described technology, which generates an accurate and efficient electromagnetic response for sensor simulators. Figure 8-1 and Figure 8-2 Compared to the results shown, Figure 8-3 The results shown demonstrate a very high correlation between simulated radar and real radar.

[0110] Example Method

[0111] Figure 9 An example method 900 for a modified ray tracker for an electromagnetic response simulator is shown. Operations (or steps) 902 to 906 are performed, but are not necessarily limited to the order or combination of operations shown herein. Furthermore, any one or more operations may be repeated, combined, or recombined to provide other operations.

[0112] At step 902, the electromagnetic sensor receives target information of a simulated target in an environment with a ground plane and sensor parameters of the electromagnetic sensor. The target information may include the target's position, orientation, and velocity. The sensor parameters may include the sensor's position, orientation, field of view, and operating frequency.

[0113] At step 904, the electromagnetic ray interacting with the target is determined based on the target information and sensor parameters. The electromagnetic ray can be determined using a modified SBR method that considers both forward and backward propagation paths according to electromagnetic mirror theory. Using the modified SBR method and electromagnetic mirror theory allows for lower complexity and higher accuracy in ray calculation.

[0114] At step 906, the electromagnetic response is modeled by approximating the electromagnetic energy reflected by the target alone or in combination with the ground plane using electromagnetic rays interacting with the target. The electromagnetic response can be approximated by an EM module that calculates the electromagnetic response based on the electromagnetic rays interacting with the target. The sensor simulator can use the electromagnetic response to accurately represent the electromagnetic energy detected by the sensor in the real world.

[0115] Figure 10 Another example method 1000 for a modified ray tracker for an electromagnetic response simulator is shown. Operations (or steps) 1002 to 1012 are performed, but are not necessarily limited to the order or combination of operations shown herein. Furthermore, any one or more operations may be repeated, combined, or recombined to provide other operations.

[0116] At step 1002, the electromagnetic sensor receives target information from multiple simulated objects in the environment and sensor parameters of the electromagnetic sensor. The target information may include the target's position, orientation, and velocity. The sensor parameters may include the sensor's position, orientation, field of view, and operating frequency.

[0117] At step 1004, sparse electromagnetic rays are emitted in the electromagnetic simulator. The rays can be emitted by a ray emitter using an adaptive ray emission process, spanning the azimuth of the electromagnetic sensor. The adaptive ray emission process emits uniformly distributed sparse rays towards targets in the simulated environment by finding the angular occupancy grid of multiple targets. The angular occupancy grid can be found by using mirror theory to determine the mirror images of multiple targets with respect to the ground plane. In this case, the ground plane can be based on electromagnetic mirror theory and defined by surfaces that contribute to the multipath reflections of the sparse electromagnetic rays as they travel to the multiple targets. The three-dimensional boundaries of each target and each corresponding target mirror image can then be determined. The sparse rays are emitted to the occupied pixels of the occupancy grid.

[0118] At step 1006, ray tracing is performed on the sparse electromagnetic rays. The sparse rays can collide and bounce between targets until the second bounce.

[0119] At step 1008, the angular ray density of the sparse electromagnetic rays reflected by each of the plurality of targets is determined. The angular ray density can be calculated by finding the length of each sparse ray up to the second bounce for each occupied pixel in the occupied grid. The length of each sparse ray is the distance each sparse ray travels to each corresponding target. The desired spatial ray density on the target or mirror surface is divided by the length of the electromagnetic ray that hits the occupied pixel representing the corresponding target or mirror to calculate the angular ray density. In one example, the desired spatial ray density could be approximately 0.01 meters. Other desired spatial ray densities can also be used.

[0120] In step 1010, dense electromagnetic rays are emitted toward multiple targets. The dense rays give distant targets a similar fidelity to closer targets.

[0121] At step 1012, the scattering of electromagnetic energy of the target is approximated by examining dense electromagnetic rays. The reflection of electromagnetic energy can then be approximated by generating an equivalent surface current for each electromagnetic ray in the dense electromagnetic rays on at least one target based on the incident field. Multiple possible scattering paths of the dense electromagnetic rays can be examined, and in response to any scattering path being the desired scattering path, the polarization scattering fields of the dense electromagnetic rays with different polarizations are calculated. The polarization scattering fields, departure direction, and arrival direction of the desired scattering path are recorded (e.g., based on the departure and arrival directions) for approximation. Furthermore, the propagation distance and time of flight (e.g., from transmitter to receiver) of each corresponding electromagnetic ray are recorded. The approximation includes simulating multipath reflection of real-world electromagnetic energy. These approximations are used to model the electromagnetic response of the electromagnetic sensor. By making targets at a distance have a similar fidelity to any closer target, the modeled electromagnetic response can accurately reflect the real-world electromagnetic energy detected by the sensor.

[0122] Figure 11 Another example method 1100 for a modified ray tracker for an electromagnetic response simulator is shown. Operations (or steps) 1102 to 1110 are performed, but are not necessarily limited to the order or combination of operations shown herein. Furthermore, any one or more operations may be repeated, combined, or recombined to provide other operations.

[0123] At step 1102, first electromagnetic ray information of electromagnetic rays simulated in the environment (e.g., a set of electromagnetic rays) is received. The first electromagnetic ray information may include the origin and direction of the ray and is relative to a global coordinate system of the environment. The first electromagnetic ray information may be stored and received by a ray tracker.

[0124] At step 1104, the closest potential target to the electromagnetic ray is determined by examining the first electromagnetic ray information. The proximity of the potential target to the electromagnetic ray is determined by a ray tracker in the global coordinate system using a pre-computed acceleration data structure containing data about the potential target. The acceleration data structure can be populated with geometric information derived from the CAD model of the potential target. Potential targets with the same CAD model (e.g., a single geometric profile) can utilize a similar acceleration data structure, and if the geometry of the potential target (e.g., a pre-computed constant geometric profile) remains static throughout the electromagnetic sensor simulation, the acceleration data structure for each potential target can remain the same for each frame of the simulation. That is, the acceleration data structure does not need to be recalculated for each frame of the simulation. Furthermore, the polarization reflection coefficient of each potential target from different incident angles can be determined based on the material of each target and stored in a polarization reflection coefficient lookup table. Additionally, for each target that is penetrable at a threshold, a transmission coefficient can be determined.

[0125] At step 1106, the first electromagnetic ray information is converted into second electromagnetic ray information. The second electromagnetic ray information is relative to the local coordinate system of the potential target.

[0126] At step 1108, it is determined whether an electromagnetic ray (e.g., a subset of electromagnetic rays) hits a face of a potential target by examining the second electromagnetic ray information. This determination is made using a ray tracing algorithm and leveraging a pre-computed, reusable acceleration data structure of the potential target. If the electromagnetic ray does not hit a face of the potential target, steps 1104 through 1108 are repeated for the next nearest potential target. These steps are repeated until a face of the potential target is determined to have been hit. If no potential target has a face that has been hit, the ray tracker determines that the electromagnetic ray did not hit the target. If a hit is determined, third electromagnetic ray information is generated for the reflected ray. The third electromagnetic information may include the bounce point (e.g., the starting point of the reflected ray) and the direction of the reflected ray. The third electromagnetic information is in a local coordinate system relative to the potential target.

[0127] At step 1110, it is determined whether to calculate the electromagnetic response of the potential target. If the surface of the potential target is hit, the electromagnetic response of the target is calculated. If the surface of the potential target is not hit, the electromagnetic response is not calculated. If the surface of the potential target is hit, an equivalent surface current based on the incident fields with different biases and the reflected electromagnetic rays induced by them can be generated. The simplified large element PO formula will be used to calculate the scattered field with up to five different paths returning to the radar from the equivalent surface current. The scattered field is calculated for different combinations of polarization of the incident and scattered waves. The scattered field information is stored in the electromagnetic rays. The third electromagnetic information can be converted into a fourth electromagnetic information relative to the global coordinate system of the environment.

[0128] Then, the fourth electromagnetic information can be treated as new first electromagnetic information, and steps 1102 to 1110 can be repeated iteratively until a stopping condition is met (none of the rays hit the target or the maximum number of iterations is reached). That is, the fourth electromagnetic information can replace the first electromagnetic ray information for the next potential target. In this way, efficient ray tracing can be performed without unnecessary calculations. Finally, electromagnetic rays with similar properties (departure angle / arrival angle, time of flight, and Doppler shift) are combined, and the total electromagnetic response of the potential target is represented by multiple clustered electromagnetic rays. The calculated electromagnetic response can then be recorded and used as an electromagnetic model for use in the execution of electromagnetic sensor simulations input to the environment.

[0129] Additional examples

[0130] Example 1: A method comprising: receiving target information of a target simulated in an environment having a ground plane and sensor parameters of an electromagnetic sensor; determining, based on the target information and sensor parameters, an electromagnetic ray interacting with the target, the electromagnetic ray having a forward propagation path to the target and a backward propagation path from the target; and modeling an electromagnetic response by approximating the electromagnetic energy reflected by the target alone or in combination with the ground plane based on the determined electromagnetic ray interacting with the target.

[0131] Example 2: The method of Example 1, wherein determining the electromagnetic rays interacting with the target includes: determining a target mirror image of the target, the target mirror image being a target mirror image with respect to a ground plane; determining an electromagnetic sensor mirror image of an electromagnetic sensor mirror image with respect to a ground plane; emitting initial electromagnetic rays from the electromagnetic sensor toward the target and the target mirror image, the initial electromagnetic rays having the same spatial ray density relative to the target and the target mirror image; performing ray tracing to determine reflected electromagnetic rays from the target and the target mirror image; and evaluating the scattered field of the reflected electromagnetic rays received by the electromagnetic sensor and the electromagnetic sensor mirror image based on equivalent surface current.

[0132] Example 3: A method in any of the preceding examples, wherein evaluating the scattered field of the reflected electromagnetic ray received by the electromagnetic sensor and the electromagnetic sensor mirror further comprises: calculating the dielectric Fresnel reflection coefficients of the target and the ground plane in response to the corresponding ray of the initial electromagnetic ray being reflected by the target or passing through the ground plane; and multiplying the scattered field and the corresponding incident field by the dielectric Fresnel reflection coefficients.

[0133] Example 4: A method in any of the preceding examples, wherein evaluating the scattered field of reflected electromagnetic rays includes: calculating the incident field of electromagnetic rays on the target using geometrical optics; and calculating the equivalent surface current on the target and the scattered field of reflected electromagnetic rays using physical optics.

[0134] Example 5: The approach in any of the previous examples, where mirror theory is used to model the electromagnetic response of the simulated target.

[0135] Example 6: The method of any of the previous examples, where the bouncing ray method is used to approximate the electromagnetic energy propagating to and being reflected by the target.

[0136] Example 7: A method in any of the previous examples, where the electromagnetic energy approximately reflected by the target includes multipath reflections from the ground plane.

[0137] Example 8: A method in any of the preceding examples, wherein the electromagnetic energy approximately reflected by the target does not include reflections determined to be blocked by the physical properties of the target.

[0138] Example 9: A method from any of the previous examples, where the target information includes: the target's position; the target's orientation; and the target's velocity.

[0139] Example 10: A method in any of the previous examples, wherein the parameters of the electromagnetic sensor include: the position of the electromagnetic sensor; the orientation of the electromagnetic sensor; the field of view of the electromagnetic sensor; and the operating frequency of the electromagnetic sensor.

[0140] Example 11: A system comprising: at least one processor configured to: receive target information of a target simulated in an environment having a ground plane and sensor parameters of an electromagnetic sensor; determine, based on the target information and sensor parameters, an electromagnetic ray interacting with the target, the electromagnetic ray having a forward propagation path to the target and a backward propagation path from the target; and model an electromagnetic response by approximating the electromagnetic energy reflected by the target alone or in combination with the ground plane based on the determined electromagnetic ray interacting with the target.

[0141] Example 12: A system in any of the preceding examples, wherein at least one processor is configured to determine electromagnetic rays associated with a target by at least the following means: determining a target mirror image of the target, the target mirror image being a target mirror image with respect to a ground plane; determining an electromagnetic sensor mirror image of an electromagnetic sensor mirror image with respect to a ground plane; emitting initial electromagnetic rays from the electromagnetic sensor toward the target and the target mirror image, the initial electromagnetic rays having the same spatial ray density relative to the target and the target mirror image; performing ray tracing to determine reflected electromagnetic rays from the target and the target mirror image; and evaluating the scattered field of the reflected electromagnetic rays received by the electromagnetic sensor and the electromagnetic sensor mirror image based on equivalent surface current.

[0142] Example 13: A system in any of the preceding examples, wherein at least one processor is configured to evaluate the scattered field of reflected electromagnetic rays received by an electromagnetic sensor and an electromagnetic sensor mirror by at least the following: calculating the dielectric Fresnel reflection coefficients of the target and the ground plane in response to the corresponding ray of the initial electromagnetic ray being reflected by the target or passing through the ground plane; multiplying the scattered field and the corresponding incident field by the dielectric Fresnel reflection coefficients; calculating the incident field of the electromagnetic ray on the target using ray tracing and geometric optics; and calculating the equivalent surface current on the target and the scattered field of the reflected electromagnetic ray using physical optics.

[0143] Example 14: A system in any of the preceding examples, wherein at least one processor is configured to approximate the electromagnetic energy reflected by the target by including multipath reflections from the ground plane.

[0144] Example 15: A system in any of the preceding examples, wherein at least one processor is configured to approximate the electromagnetic energy reflected by the target by excluding reflections determined to be obscured by the physical properties of the target.

[0145] Example 16: A computer-readable storage medium comprising instructions that, when executed, cause at least one processor to: receive target information of a target simulated in an environment having a ground plane and sensor parameters of an electromagnetic sensor; determine, based on the target information and sensor parameters, an electromagnetic ray interacting with the target, the electromagnetic ray having a forward propagation path to the target and a backward propagation path from the target; and model an electromagnetic response by approximating electromagnetic energy reflected by the target alone or in combination with the ground plane based on the determined electromagnetic ray interacting with the target.

[0146] Example 17: A computer-readable storage medium of any of the preceding examples, wherein the instructions, when executed, further cause at least one processor to determine electromagnetic rays interacting with a target by at least the following means: determining a target mirror image of the target, the target mirror image being a target mirror image with respect to a ground plane; determining an electromagnetic sensor mirror image of an electromagnetic sensor mirror image with respect to a ground plane; emitting initial electromagnetic rays from the electromagnetic sensor toward the target and the target mirror image, the initial electromagnetic rays having the same spatial ray density relative to the target and the target mirror image; performing ray tracing to determine reflected electromagnetic rays from the target and the target mirror image; and evaluating the scattered field of the reflected electromagnetic rays received by the electromagnetic sensor and the electromagnetic sensor mirror image based on equivalent surface current.

[0147] Example 18: A computer-readable storage medium of any of the preceding examples, wherein the instructions, when executed, further cause at least one processor to evaluate the scattered field of reflected electromagnetic rays received by an electromagnetic sensor and an electromagnetic sensor mirror by at least the following: calculating the dielectric Fresnel reflection coefficients of the target and the ground plane in response to the corresponding ray of the initial electromagnetic ray being reflected by the target or passing through the ground plane; multiplying the scattered field and the corresponding incident field by the dielectric Fresnel reflection coefficients; calculating the incident field of the electromagnetic ray on the target using ray tracing and geometric optics; and calculating the equivalent surface current on the target and the scattered field of the reflected electromagnetic ray using physical optics.

[0148] Example 19: A computer-readable storage medium of any of the preceding examples, wherein instructions, when executed, cause at least one processor to approximate electromagnetic energy reflected by a target by means of multipath reflections, including from the ground plane.

[0149] Example 20: A computer-readable storage medium of any of the preceding examples, wherein the instructions, when executed, cause at least one processor to approximate electromagnetic energy reflected by a target by means of reflections that are determined to be obscured by the physical characteristics of the target.

[0150] Example 21: A method comprising: receiving sensor parameters (including a field of view) of an electromagnetic sensor simulated in an environment and target information of one or more targets in the field of view; emitting sparse electromagnetic rays in the field of view; determining, in response to the sparse electromagnetic rays reflected from the one or more targets, the angular ray density of a dense electromagnetic ray to be emitted toward the one or more targets; and modeling the electromagnetic response of the electromagnetic sensor based on approximating the reflection of electromagnetic energy from the targets, taking into account multipath reflections, by emitting the dense electromagnetic rays toward the one or more targets.

[0151] Example 22: The method of Example 21, where the sparse rays are uniformly distributed.

[0152] Example 23: A method in any of the preceding examples, wherein emitting sparse electromagnetic rays in the field of view comprises: generating a corner grid of pixels of the environment included in the field of view based on the field of view of an electromagnetic sensor; determining, based on target information, pixels in the corner grid of pixels occupied by each of one or more targets; and emitting sparse electromagnetic rays to the center of each pixel in the corner grid of pixels.

[0153] Example 24: A method for any of the previous examples, where the resolution of the corner grid of the pixel is approximately 0.1 degrees.

[0154] Example 25: A method in any of the preceding examples, wherein determining pixels in a corner grid occupied by each of the one or more targets comprises: determining the mirror image of the one or more targets about a ground plane defined by a surface contributing to multipath reflection of electromagnetic radiation, based on mirror theory; and determining the three-dimensional boundary of each of the one or more targets and the mirror image of each of the one or more targets, wherein the one or more targets further include the mirror image of the one or more targets.

[0155] Example 26: A method in any of the preceding examples, wherein determining the angular ray density of dense electromagnetic rays includes: calculating the length of sparse electromagnetic rays reflected from one or more targets based on the distance traveled by sparse electromagnetic rays to reach the corresponding target.

[0156] Example 27: A method in any of the preceding examples, wherein the length of the sparse electromagnetic ray comprises: a first distance from the electromagnetic sensor to the first reflection point; and a second distance from the first reflection point to the second reflection point.

[0157] Example 28: A method of any of the preceding examples further includes: dividing the desired spatial ray density by the length of the sparse electromagnetic rays reflected from one or more targets.

[0158] Example 29: A method from any of the previous examples, where the desired spatial ray density is approximately 0.01 meters.

[0159] Example 30: A method in any of the previous examples, wherein the distance a sparse ray travels to reach the corresponding target does not include sparse electromagnetic rays with more than one multipath reflection.

[0160] Example 31: A method in any of the preceding examples, wherein the multipath reflection considered includes a plurality of possible scattering paths, including: a first scattering path comprising a first forward propagation path to the multipath reflection, a first target reflection, and a first reverse propagation path directly back to the electromagnetic sensor; a second scattering path comprising a first forward propagation path to the multipath reflection, a second target reflection, and a second reverse propagation path returning to a mirror image of the electromagnetic sensor; a third scattering path comprising a first forward propagation path to the multipath reflection, a third target reflection, and a third reverse propagation path traveling in the opposite direction to the first forward propagation path; a fourth scattering path being the reverse path of the first scattering path; and a fifth scattering path being the reverse path of the second scattering path.

[0161] Example 32: A method in any of the preceding examples, wherein the reflection of approximate electromagnetic energy includes: receiving geometric information of dense electromagnetic rays; calculating the equivalent surface current on the corresponding target using physical optics based on incident fields with different polarizations; examining possible scattering paths of the dense electromagnetic rays; calculating the polarization scattering field of the dense electromagnetic rays with different polarizations based on the desired scattering path using physical optics in response to one or more of the possible scattering paths being the desired scattering path; and recording the polarization scattering field, the propagation distance and flight time of each corresponding electromagnetic ray of the dense electromagnetic rays from the transmitter to the receiver of the electromagnetic sensor, and the departure and arrival directions of each of the desired scattering paths.

[0162] Example 33: A system comprising: at least one processor configured to: receive sensor parameters (including a field of view) of an electromagnetic sensor simulated in an environment and target information of one or more targets in the field of view; emit sparse electromagnetic rays in the field of view; determine the angular ray density of a dense electromagnetic ray to be emitted toward the one or more targets in response to the sparse electromagnetic rays reflected from the one or more targets; and model the electromagnetic response of the electromagnetic sensor by approximating the reflection of electromagnetic energy from the targets, taking into account multipath reflections, based on the emission of the dense electromagnetic rays toward the one or more targets.

[0163] Example 34: A system in any of the preceding examples, wherein at least one processor is configured to emit sparse electromagnetic rays in a field of view at least by: generating a corner grid of pixels comprising the environment in the field of view based on the field of view of an electromagnetic sensor; determining pixels in the corner grid of pixels occupied by each of one or more targets based on target information; and emitting sparse electromagnetic rays to the center of each pixel in the corner grid of pixels.

[0164] Example 35: A system in any of the preceding examples, wherein at least one processor is configured to determine pixels in a corner grid of pixels occupied by each of one or more targets by at least the following: determining the mirror image of one or more targets about a ground plane defined by a surface contributing to multipath reflection of electromagnetic radiation, based on mirror theory; and determining the three-dimensional boundary of each of the one or more targets and the mirror image of each of the mirror images of one or more targets, wherein the one or more targets further include the mirror image of one or more targets.

[0165] Example 36: A system in any of the preceding examples, wherein at least one processor is configured to determine the angular ray density of dense electromagnetic rays by calculating the length of sparse electromagnetic rays reflected from one or more targets based on the distance traveled by sparse electromagnetic rays to reach the corresponding targets.

[0166] Example 37: A system in any of the preceding examples, wherein the length of the sparse electromagnetic ray includes: a first distance from the electromagnetic sensor to the first reflection point; and a second distance from the first reflection point to the second reflection point.

[0167] Example 38: A system in any of the preceding examples, wherein at least one processor is further configured to determine the angular ray density of dense electromagnetic rays by dividing the desired spatial ray density by the length of sparse electromagnetic rays reflected from one or more targets.

[0168] Example 39: A computer-readable storage medium comprising instructions that, when executed, cause at least one processor to: receive sensor parameters (including a field of view) of an electromagnetic sensor simulated in an environment and target information of one or more targets in the field of view; emit sparse electromagnetic rays in the field of view; determine the angular ray density of a dense electromagnetic ray to be emitted toward the one or more targets in response to the sparse electromagnetic rays reflected from the one or more targets; and model the electromagnetic response of the electromagnetic sensor by approximating the reflection of electromagnetic energy from the targets, taking into account multipath reflections, based on the emission of the dense electromagnetic rays toward the one or more targets.

[0169] Example 40: A computer-readable storage medium of any of the preceding examples, the computer-readable storage medium comprising instructions that, when executed, cause at least one processor to emit sparse electromagnetic rays in a field of view at least by: generating a corner grid of pixels comprising the environment in the field of view based on the field of view of an electromagnetic sensor; determining, based on target information, pixels in the corner grid of pixels occupied by each of one or more targets; and emitting sparse electromagnetic rays to the center of each pixel in the corner grid of pixels.

[0170] Example 41: A method comprising: receiving first electromagnetic ray information of electromagnetic rays simulated in an environment including a plurality of potential targets, the first electromagnetic ray information including the starting point and orientation of the electromagnetic ray relative to a global coordinate system of the environment; determining the potential target closest to the first electromagnetic ray information among the plurality of potential targets; converting the first electromagnetic ray information into second electromagnetic ray information based on the potential target, the second electromagnetic ray information including the starting point and orientation of the electromagnetic ray relative to a local coordinate system of the potential target; determining whether the electromagnetic ray hits a surface of the potential target based on the second electromagnetic ray information and a pre-computed acceleration data structure indicating the geometric contour of the potential target; and determining whether to compute an electromagnetic response of the potential target based on whether the electromagnetic ray hits a surface of the potential target.

[0171] Example 42: The method of Example 41, wherein determining whether to calculate the electromagnetic response of a potential target includes: determining to calculate the electromagnetic response of the potential target in response to determining that an electromagnetic ray hits the surface of the potential target; and outputting the electromagnetic response to an electromagnetic model in response to calculating the electromagnetic response for use in the execution of an electromagnetic sensor simulation input to the environment.

[0172] Example 43: A method in any of the preceding examples, wherein the calculation of the electromagnetic response includes: determining a reflected electromagnetic ray reflected from a potential target based on an electromagnetic ray striking the surface of the potential target; determining third electromagnetic ray information, the third electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to the local coordinate system of the potential target; and converting the third electromagnetic ray information into fourth electromagnetic ray information, the fourth electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to the global coordinate system of the environment, the fourth electromagnetic ray information replacing the received first electromagnetic ray information for the next potential target.

[0173] Example 44: A method in any of the preceding examples, wherein the calculation of the electromagnetic response further includes: generating an equivalent surface current induced by the incident field based on the received first electromagnetic information, based on an incident field with different polarizations; examining multiple possible scattering paths of the reflected electromagnetic ray; calculating the scattering field of the reflected electromagnetic ray with different polarizations based on the desired scattering path, in response to one or more of the multiple possible scattering paths being the desired scattering path of the reflected electromagnetic ray; and recording the scattering field, departure direction, and arrival direction of each scattering path in the desired scattering path of the reflected electromagnetic ray.

[0174] Example 45: A method in any of the preceding examples, wherein determining whether to compute the electromagnetic response of a potential target includes: determining not to compute the electromagnetic response of the potential target in response to determining that an electromagnetic ray did not hit a surface of the potential target; and the method further includes: determining another potential target among a plurality of potential targets based on first electromagnetic ray information; converting the first electromagnetic ray information into another second electromagnetic ray information based on the other potential target, the other second electromagnetic ray information including a starting point and orientation relative to a local coordinate system of the other potential target; determining whether the electromagnetic ray hit a surface of the other potential target based on the other second electromagnetic ray information; and determining whether to compute the electromagnetic response associated with the electromagnetic ray based on the other potential target based on whether the electromagnetic ray hit a surface of the other potential target.

[0175] Example 46: A method for any of the previous examples, where another potential target is the next closest potential target.

[0176] Example 47: A method from any of the previous examples, where a pre-computed acceleration data structure represents a potential target over the duration of an electromagnetic sensor simulation.

[0177] Example 48: A method in any of the previous examples, where the same pre-computed acceleration data structure indicating a single geometric profile is used to represent multiple potential targets with the same computer-aided design (CAD) model.

[0178] Example 49: The method of any of the preceding examples further includes: determining the polarization reflection coefficient of each of the one or more potential targets based on the material properties of each of the one or more potential targets.

[0179] Example 50: A method in any of the preceding examples further includes: generating a polarization reflection coefficient lookup table, the lookup table including the polarization reflection coefficient of each of one or more potential targets.

[0180] Example 51: The method of any of the preceding examples further includes: determining the transmission coefficient of each of the one or more potential targets based on the fact that the material of each of the one or more potential targets is permeable at a threshold.

[0181] Example 52: A system comprising: at least one processor configured to: receive first electromagnetic ray information of a simulated set of electromagnetic rays in an environment including a plurality of potential targets, the first electromagnetic ray information including the starting point and orientation of each corresponding electromagnetic ray relative to a global coordinate system of the environment; determine the potential target among the plurality of potential targets that is closest to the first electromagnetic ray information; convert the first electromagnetic ray information into second electromagnetic ray information based on the potential target, the second electromagnetic ray information including the starting point and orientation relative to a local coordinate system of the potential target; and determine, based on the second electromagnetic ray information and a pre-computed acceleration data structure indicating the geometric contours of the potential target, a... Whether a subset of electromagnetic rays hits a surface of a potential target; in response to determining that a subset of electromagnetic rays hits a surface of a potential target, calculating the electromagnetic response of the subset of electromagnetic rays hitting the surface of the potential target; determining whether a subset of electromagnetic rays hits any surface of the potential target; and in response to determining that a subset of electromagnetic rays hits a surface of the potential target: calculating the electromagnetic response of the subset of electromagnetic rays hitting the surface of the potential target; and outputting the electromagnetic response to an electromagnetic model for use in the execution of an electromagnetic sensor simulation input to the environment; or in response to determining that a subset of electromagnetic rays does not hit any surface of the potential target, avoiding the calculation or output of the electromagnetic response of the subset of electromagnetic rays that does not hit any surface of the potential target.

[0182] Example 53: The system of Example 52, wherein at least one processor is further configured to, in response to determining that a subset of electromagnetic rays does not hit any face of a potential target, further: determine another potential target among a plurality of potential targets, the other potential target being the next closest potential target to each electromagnetic ray in the set of electromagnetic rays at a corresponding starting point and direction; convert first electromagnetic ray information into another second electromagnetic ray information based on the other potential target, the other second electromagnetic ray information including the starting point and direction relative to the local coordinate system of the other potential target; determine whether a subset of electromagnetic rays hits a face of the other potential target based on the other second electromagnetic ray information; and calculate and output the electromagnetic response of the other potential target to the model based on whether a subset of electromagnetic rays hits a face of the other potential target.

[0183] Example 54: A system in any of the preceding examples, wherein at least one processor is further configured to calculate the electromagnetic response by at least the following manner: determining at least one reflected electromagnetic ray reflected from the potential target based on a subset of electromagnetic rays striking the surface of the potential target; determining third electromagnetic ray information, the third electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to the local coordinate system of the potential target; and converting the third electromagnetic ray information into fourth electromagnetic ray information, the fourth electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to the global coordinate system of the environment, the fourth electromagnetic ray information replacing the received first electromagnetic ray information for the next potential target.

[0184] Example 55: A system in any of the preceding examples, wherein at least one processor is further configured to: generate an equivalent surface current induced by the incident field based on first electromagnetic information, based on an incident field having different polarizations; examine multiple possible scattering paths of the reflected electromagnetic ray; in response that one or more of the multiple possible scattering paths is a desired scattering path of the reflected electromagnetic ray, calculate the scattering field of the reflected electromagnetic ray with different polarizations based on the desired scattering path; and record the scattering field, departure direction, and arrival direction of each scattering path of the desired scattering path of the reflected electromagnetic ray.

[0185] Example 56: A system in any of the previous examples, where a pre-computed acceleration data structure represents a potential target over the duration of an electromagnetic sensor simulation.

[0186] Example 57: A system in any of the preceding examples, wherein at least one processor is further configured to use a pre-computed acceleration data structure to represent multiple potential targets having the same computer-aided design (CAD) model by using the same pre-computed acceleration data structure indicating a single geometric profile.

[0187] Example 58: A system in any of the preceding examples, wherein at least one processor is further configured to: determine the polarization reflectance coefficient of each of the one or more potential targets based on the material properties of each of the one or more potential targets; generate a polarization reflectance coefficient lookup table including the polarization reflectance coefficient of each of the one or more potential targets; and determine the transmission coefficient of each of the one or more potential targets based on the material of each of the one or more potential targets being transmissible at a threshold.

[0188] Example 59: A computer-readable storage medium comprising instructions that, when executed, cause at least one processor to: receive first electromagnetic ray information of a set of electromagnetic rays simulated in an environment including a plurality of potential targets, the first electromagnetic ray information including the starting point and direction of each corresponding electromagnetic ray relative to a global coordinate system of the environment; determine the potential target among the plurality of potential targets that is closest to the first electromagnetic ray information; convert the first electromagnetic ray information into second electromagnetic ray information based on the potential target, the second electromagnetic ray information including the starting point and direction relative to a local coordinate system of the potential target; and based on the second electromagnetic ray information and a pre-calculated acceleration indicating the geometric contour of the potential target. The data structure determines whether a subset of electromagnetic rays hits a surface of a potential target; in response to determining that a subset of electromagnetic rays hits a surface of a potential target, it calculates the electromagnetic response of the subset of electromagnetic rays hitting the surface of the potential target; it determines whether a subset of electromagnetic rays hits any surface of the potential target; and in response to determining that a subset of electromagnetic rays hits a surface of the potential target, it calculates the electromagnetic response of the subset of electromagnetic rays hitting the surface of the potential target; and it outputs the electromagnetic response to an electromagnetic model for use in the execution of an electromagnetic sensor simulation input to the environment; or in response to determining that a subset of electromagnetic rays does not hit any surface of the potential target, it avoids calculating or outputting the electromagnetic response of the subset of electromagnetic rays that does not hit any surface of the potential target.

[0189] Example 60: A computer-readable storage medium of any of the preceding examples, wherein the instructions, when executed, cause at least one processor to calculate an electromagnetic response by at least the following manner: determining at least one reflected electromagnetic ray reflected from a potential target based on a subset of electromagnetic rays striking a surface of the potential target; determining third electromagnetic ray information, the third electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to a local coordinate system of the potential target; and converting the third electromagnetic ray information into fourth electromagnetic ray information, the fourth electromagnetic ray information including the starting point and direction of the reflected electromagnetic ray relative to a global coordinate system of the environment, the fourth electromagnetic ray information replacing the received first electromagnetic ray information for the next potential target.

[0190] Conclusion

[0191] While various embodiments of the present disclosure have been described in the foregoing description and illustrated in the accompanying drawings, it should be understood that the present disclosure is not limited thereto, but can be practiced in various ways within the scope of the following claims. It will be apparent from the foregoing description that various modifications can be made without departing from the spirit and scope of the present disclosure as defined by the following claims. Problems associated with electromagnetic response in simulated environments may occur in other systems. Therefore, although described as a way to improve the electromagnetic response of a radar simulator, the techniques described above can be applied to other systems simulating other electromagnetic sensors.

[0192] Unless the context explicitly states otherwise, the use of "or" and grammatically related terms indicates an unrestricted, non-exclusive alternative. As used herein, the phrase referring to "at least one" of a list of items means any combination of those items, including a single member. As an example, "at least one of a, b, or c" is intended to cover: a, b, c, ab, ac, bc, and abc, as well as any combination with multiple identical elements (e.g., aa, aaa, aab, aac, abb, acc, bb, bbb, bbb, cc, and ccc, or any other ordering of a, b, and c).

Claims

1. A method, the method comprising: Receive sensor parameters, including the field of view, from an electromagnetic sensor simulated in the environment, as well as target information of one or more targets in the field of view; A first electromagnetic ray of a first density is emitted within the field of view; Based on the received first electromagnetic rays reflected from the one or more targets, determine the angular ray density of a second electromagnetic ray with a second density to be emitted toward the one or more targets. Where the second density is greater than the first density; and The electromagnetic response of the electromagnetic sensor is modeled based on the reflection of electromagnetic energy from the targets, approximating the reflection of electromagnetic rays emitted toward the one or more targets, taking into account multipath reflection. The reflection of the electromagnetic energy approximation includes approximations of the polarized scattering field, the propagation distance and flight time of each corresponding electromagnetic ray of the first electromagnetic ray from the transmitter to the receiver of the electromagnetic sensor, and the departure and arrival directions of each desired scattering path.

2. The method according to claim 1, characterized in that, The first electromagnetic ray is uniformly distributed.

3. The method according to claim 1, characterized in that, Emitting the first electromagnetic ray in the field of view includes: Based on the field of view of the electromagnetic sensor, a corner grid of pixels of the environment included in the field of view is generated; Based on the target information, determine the pixel in the corner grid of the pixel occupied by each of the one or more targets; and The first electromagnetic ray is emitted toward the center of each pixel in the corner grid of the pixel.

4. The method according to claim 3, characterized in that, The resolution of the corner grid of the pixel is approximately 0.1 degrees.

5. The method according to claim 3, characterized in that, Determining the pixels in the corner grid occupied by each of the one or more targets includes: Based on mirror theory, determine the mirror image of the one or more targets with respect to a ground plane, said ground plane being defined by a surface that contributes to multipath reflection of electromagnetic radiation; and Determine the three-dimensional boundaries of each of the one or more targets and the mirror image of each of the one or more targets, wherein the one or more targets also include the mirror images of the one or more targets.

6. The method according to claim 3, characterized in that, Determining the angular ray density of the second electromagnetic ray includes: The length of the first electromagnetic ray reflected from the one or more targets is calculated based on the distance the first electromagnetic ray travels to reach the corresponding target.

7. The method according to claim 6, characterized in that, The length of the first electromagnetic ray includes: The first distance from the electromagnetic sensor to the first reflection point; and The second distance from the first reflection point to the second reflection point.

8. The method according to claim 6, further comprising: The angular ray density is calculated by dividing the spatial ray density by the length of the first electromagnetic ray reflected from the one or more targets.

9. The method according to claim 8, characterized in that, The spatial ray density is approximately 0.01 meters.

10. The method according to claim 6, characterized in that, The first electromagnetic ray travels the distance to reach the corresponding target, excluding the first electromagnetic ray with more than one multipath reflection.

11. The method according to claim 1, characterized in that, The multipath reflections considered include multiple possible scattering paths, including: The first scattering path includes a first forward propagation path to multipath reflection, a first target reflection, and a first reverse propagation path directly returning to the electromagnetic sensor; The second scattering path includes a first forward propagation path to the multipath reflection, a second target reflection, and a second reverse propagation path returning to the image of the electromagnetic sensor. The third scattering path includes the first forward propagation path to the multipath reflection, the third target reflection, and the third reverse propagation path traveling in the opposite direction to the first forward propagation path. A fourth scattering path, which is the reverse path of the first scattering path; and The fifth scattering path is the inverse of the second scattering path.

12. The method according to claim 11, characterized in that, The reflection of the electromagnetic energy approximation further includes: Receive the geometric information of the second electromagnetic ray; Based on incident fields with different polarizations, the equivalent surface current on the corresponding target is calculated using physical optics. Examine the possible scattering paths of the second electromagnetic ray; In response to one or more of the possible scattering paths being the desired scattering path, based on the desired scattering path, the polarization scattering field of the second electromagnetic ray with different polarizations is calculated using physical optics; and Record the polarization scattering field, the propagation distance and flight time of each corresponding electromagnetic ray of the first electromagnetic ray from the transmitter of the electromagnetic sensor to the receiver, and the departure and arrival directions of each scattering path in the desired scattering path.

13. A system comprising: At least one processor, said at least one processor being configured to: Receive sensor parameters, including the field of view, from an electromagnetic sensor simulated in the environment, as well as target information of one or more targets in the field of view; A first electromagnetic ray of a first density is emitted within the field of view; Based on the received first electromagnetic rays reflected from the one or more targets, determine the angular density of a second electromagnetic ray with a second density to be emitted toward the one or more targets. Where the second density is greater than the first density; and The electromagnetic response of the electromagnetic sensor is modeled based on the reflection of electromagnetic energy from the targets, approximating the reflection of electromagnetic rays emitted toward the one or more targets, taking into account multipath reflection. The reflection of the electromagnetic energy approximation includes approximations of the polarized scattering field, the propagation distance and flight time of each corresponding electromagnetic ray of the first electromagnetic ray from the transmitter to the receiver of the electromagnetic sensor, and the departure and arrival directions of each desired scattering path.

14. The system according to claim 13, characterized in that, The at least one processor is configured to emit the first electromagnetic ray in the field of view in at least the following manner: Based on the field of view of the electromagnetic sensor, a corner grid of pixels of the environment included in the field of view is generated; Based on the target information, determine the pixel in the corner grid of the pixel occupied by each of the one or more targets; and The first electromagnetic ray is emitted toward the center of each pixel in the corner grid of the pixel.

15. The system according to claim 14, characterized in that, The at least one processor is configured to determine, at least in the following manner, pixels in the corner grid of pixels occupied by each of the one or more targets: The mirror image of the one or more targets with respect to the ground plane is determined based on the mirror theory, the ground plane being defined by a surface that contributes to the multipath reflection of electromagnetic radiation; as well as Determine the three-dimensional boundaries of each of the one or more targets and the mirror image of each of the one or more targets, wherein the one or more targets also include the mirror images of the one or more targets.

16. The system according to claim 14, characterized in that, The at least one processor is configured to determine the angular density of the second electromagnetic ray by at least the following means: The length of the first electromagnetic ray reflected from the one or more targets is calculated based on the distance the first electromagnetic ray travels to reach the corresponding target.

17. The system according to claim 16, characterized in that, The length of the first electromagnetic ray includes: The first distance from the electromagnetic sensor to the first reflection point; and The second distance from the first reflection point to the second reflection point.

18. The system according to claim 16, characterized in that, The at least one processor is further configured to calculate the angular ray density of the second electromagnetic ray in the following manner: Divide the desired spatial ray density by the length of the first electromagnetic ray reflected from the one or more targets.

19. A non-transient computer-readable storage medium, the computer-readable storage medium comprising instructions that, when executed, cause at least one processor to: Receive sensor parameters, including the field of view, from an electromagnetic sensor simulated in the environment, as well as target information of one or more targets in the field of view; A first electromagnetic ray of a first density is emitted within the field of view; Based on the received first electromagnetic rays reflected from the one or more targets, determine the angular ray density of a second electromagnetic ray with a second density to be emitted toward the one or more targets. Where the second density is greater than the first density; and The electromagnetic response of the electromagnetic sensor is modeled based on the reflection of electromagnetic energy from the targets, approximating the reflection of electromagnetic rays emitted toward the one or more targets, taking into account multipath reflection. The reflection of the electromagnetic energy approximation includes approximations of the polarized scattering field, the propagation distance and flight time of each corresponding electromagnetic ray of the first electromagnetic ray from the transmitter to the receiver of the electromagnetic sensor, and the departure and arrival directions of each desired scattering path.

20. The non-transient computer-readable storage medium of claim 19, further comprising instructions that, when executed, cause at least one processor to emit the first electromagnetic ray in the field of view in at least the following manner: Based on the field of view of the electromagnetic sensor, a corner grid of pixels of the environment included in the field of view is generated; Based on the target information, determine the pixel in the corner grid of the pixel occupied by each of the one or more targets; and The first electromagnetic ray is emitted toward the center of each pixel in the corner grid of the pixel.

Images