A system and method are disclosed for offering a trainee presentation blending live and virtual entities to create a training scenario unconstrained by live entity operational performance and geographical limitations. The system blends an instance of a virtual entity with an actual presentation of a live entity within a trainee presentation. Outside of trainee local sensor range, the system presents a virtual entity to the trainee while occluding local sensor presentation of the live entity. As the scenario progresses to a lessor range or higher criticality, the system offers the live entity emulation information concerning characteristics of the virtual entity so the live entity may anticipate and begin to emulate the virtual. At a crossover point, the system determines if the live entity has successfully emulated the virtual and if so, discontinues presentation of the virtual while removing the occlusion allowing presentation of the live entity.

Aspects of the disclosure relate to generating simulated degraded sensor data. For instance, first sensor data collected by a sensor of a perception system of an autonomous vehicle may be received. The first sensor data may be inputted into simulated degraded sensor data for a particular degrading condition. The simulated degraded sensor data may be used to evaluate or train a model for detecting objects of the perception system.

A hardware in the loop simulation and test system that includes a phased array antenna simulation system providing dynamic range and angle of arrival signals simulation and synchronizing for input into a system under test (SUT) that includes a phased array signal processing system along with related methods. Embodiments include system elements that increase precision of signal simulation to include reduced error in angular resolution.

The disclosure relates to a simulation device for motor vehicle monitoring, wherein a radar sensor (2) and a camera sensor (3) and a LiDAR light receiving sensor (1) and a computer (4) are present, wherein the radar sensor (2) can be controlled via a radar signal transmitter, and the camera sensor (3) can be controlled via a lens, and the LiDAR light receiving sensor (1) can be controlled via a light transmitter.

A method and system for computer-implemented simulation of radar raw data, where the radar raw data are generated for a synthetic MIMO radar system including a transmitter array of several transmitters for transmitting radar signals and a receiver array of several receivers for receiving radar echoes of the radar signals. In this method, ray tracing of a radar signal sent from a preset transmitting position within the transmitter array and received at a preset receiving position within the receiver array is performed based on a 3D model of a virtual area adjacent to the MIMO radar system, where the ray tracing determines propagations of a plurality of rays within the radar signal from the preset transmitting position to the preset receiving position. The propagation of each ray is dependent on a first angle and a second angle describing the direction of a respective ray at the preset transmitting position. By using first-order derivatives with respect to the first angle and the second angle, propagations of a plurality of modified rays originating from a respective transmitter and received at a respective receiver are determined based on a linear approximation. The modified rays are processed in order to determine the radar raw data.

A method and system for computer-implemented simulation of radar raw data, where the radar raw data are generated for a synthetic MIMO radar system including a transmitter array of several transmitters for transmitting radar signals and a receiver array of several receivers for receiving radar echoes of the radar signals. In this method, ray tracing of a radar signal sent from a preset transmitting position within the transmitter array and received at a preset receiving position within the receiver array is performed based on a 3D model of a virtual area adjacent to the MIMO radar system, where the ray tracing determines propagations of a plurality of rays within the radar signal from the preset transmitting position to the preset receiving position. The propagation of each ray is dependent on a first angle and a second angle describing the direction of a respective ray at the preset transmitting position. By using first-order derivatives with respect to the first angle and the second angle, propagations of a plurality of modified rays originating from a respective transmitter and received at a respective receiver are determined based on a linear approximation. The modified rays are processed in order to determine the radar raw data.

A method for building a coherent radar cross-section (RCS) model database for real-time dynamic simulation of range-Doppler radars is disclosed. The database may be used with radar sensors that employ different waveforms. A pre-processing operation before the dynamic simulation performs fast Fourier Transforms (FFTs) to interpolate the target frequency responses from the database to match the frequency samplings of the radar used in the dynamic simulation. The method determines the frequency responses of the targets to a reference chirp in a coherent processing interval (CPI) and the radial velocities of the targets relative to the radar at the time of the reference chirp. The method extrapolates, using FFTs, the frequency responses of the targets to the reference chirp across the velocity dimension based on the relative radial velocities to determine the frequency responses of the targets to the other chirps across the CPI, reducing the computational burden for the simulation.

A method and system for computer-implemented simulation of radar raw data, where the radar raw data are generated for a synthetic MIMO radar system including a transmitter array of several transmitters for transmitting radar signals and a receiver array of several receivers for receiving radar echoes of the radar signals. In this method, ray tracing of a radar signal sent from a preset transmitting position within the transmitter array and received at a preset receiving position within the receiver array is performed based on a 3D model of a virtual area adjacent to the MIMO radar system, where the ray tracing determines propagations of a plurality of rays within the radar signal from the preset transmitting position to the preset receiving position. The propagation of each ray is dependent on a first angle and a second angle describing the direction of a respective ray at the preset transmitting position. By using first-order derivatives with respect to the first angle and the second angle, propagations of a plurality of modified rays originating from a respective transmitter and received at a respective receiver are determined based on a linear approximation. The modified rays are processed in order to determine the radar raw data.

A method and system for computer-implemented simulation of radar raw data, where the radar raw data are generated for a synthetic MIMO radar system including a transmitter array of several transmitters for transmitting radar signals and a receiver array of several receivers for receiving radar echoes of the radar signals. In this method, ray tracing of a radar signal sent from a preset transmitting position within the transmitter array and received at a preset receiving position within the receiver array is performed based on a 3D model of a virtual area adjacent to the MIMO radar system, where the ray tracing determines propagations of a plurality of rays within the radar signal from the preset transmitting position to the preset receiving position. The propagation of each ray is dependent on a first angle and a second angle describing the direction of a respective ray at the preset transmitting position. By using first-order derivatives with respect to the first angle and the second angle, propagations of a plurality of modified rays originating from a respective transmitter and received at a respective receiver are determined based on a linear approximation. The modified rays are processed in order to determine the radar raw data.

A method of detecting obstacle vehicles present in a virtual driving environment by using a virtual radar sensor for an ADAS test of a vehicle is disclosed. The disclosed obstacle detection method may include: establishing an obstacle vehicle candidate group from at least one obstacle vehicles each represented by four points in a virtual driving environment, where the obstacle vehicle candidate group includes obstacle vehicles that are wholly or partially included in a sensing range of the virtual radar sensor; updating the obstacle vehicle candidate group by excluding an obstacle vehicle that is located in a shadow region from the obstacle vehicle candidate group; and calculating the shortest distance between an obstacle vehicle included in the updated obstacle vehicle candidate group and the virtual radar sensor.