Systems and methods are provided for generating a radar model for a target object. In embodiments, a target simulation model is received that represents one or more physical aspects of a target object, an environment simulation model is received that represents one or more physical aspects of an environment object, and a target distance parameter is received that identifies a reference distance between the target object and a radar system to be simulated. A simulation model is generated based, at least in part, on the target simulation model, the environment simulation model, and the reference distance, and further based on a target aspect angle that identifies an angular position of the target object in relation to the radar system. Interaction of the radar system with the target object and the environment object is simulated using the simulation model, and results of the simulation are used to generate a range profile for the target object at the target aspect angle, wherein the range profile identifies a radar return strength for the reference distance. The target aspect angle is then incremented, and the operations are repeated until range profiles are generated for the target object at a plurality of target angles amounting to a 360 degree rotation of the target object. The range profiles at the plurality of target angles are then accumulated to generate the radar model for the target object.

Simulated degraded sensor data may be generated for use in training a model. For instance, first sensor data collected by a sensor of a perception system of an autonomous vehicle may be received and converted into the simulated degraded sensor data for a particular degrading condition, such as a weather-related degrading condition. Then, the simulated degraded sensor data may be used to train a model for evaluating performance of the perception system to detect objects external to the autonomous vehicle under one or more conditions.

This document describes techniques and systems for accurate and efficient electromagnetic response for a sensor simulator. Target information and sensor parameters for an electromagnetic sensor are simulated in an environment that includes a ground plane. Electromagnetic rays that may be detected by the sensor or an image of the sensor are launched from the simulated sensor toward the target and an image of the target about the ground plane to determine a complex electromagnetic response of the target. A ray-tracing algorithm is applied to trace the forward wave propagation of electromagnetic rays in the environment that considers rays bouncing between the target and the image of the target. An electromagnetic response can be modeled based on the congregation of the electromagnetic response of all backward paths of all bounces of all rays. In this manner, an efficient and accurate electromagnetic response model may be approximated.

This document describes techniques and systems for accurate and efficient electromagnetic response for a sensor simulator. Target information and sensor parameters for an electromagnetic sensor are simulated in an environment that includes a ground plane. Electromagnetic rays that may be detected by the sensor or an image of the sensor are launched from the simulated sensor toward the target and an image of the target about the ground plane to determine a complex electromagnetic response of the target. A ray-tracing algorithm is applied to trace the forward wave propagation of electromagnetic rays in the environment that considers rays bouncing between the target and the image of the target. An electromagnetic response can be modeled based on the congregation of the electromagnetic response of all backward paths of all bounces of all rays. In this manner, an efficient and accurate electromagnetic response model may be approximated.

Systems and methods for generating simulated LiDAR data using RADAR and image data are provided. An algorithm is trained using deep-learning techniques such as loss functions to generate simulated LiDAR data using RADAR and image data. Once trained, the algorithm can be implemented in a system, such as a vehicle, equipped with RADAR and image sensors in order to generate simulated LiDAR data describing the system's environment. The simulated LiDAR data may be used by a vehicle control system to determine, generate, and implement modified driving operations.

A training simulation system and method for detection of hazardous materials simulates real-world hazardous environments to provide a trainee with hazardous material training. The system provides a hazardous material detection simulator that displays simulated readings to indicate presence thereof. The detection simulator automatically generates the simulated readings, based on its relative position to the hazard point, and based on preprogrammed hazard points in the area. A host trainer, through a trainer communication device, remotely generates and adjusts the simulated readings while tracking vehicle's position. A vehicle integrally contains the hazardous material detection simulator. A trainee controls the vehicle while also observing and reacting to the simulated readings. Once the hazard point is determined, based on simulated readings, the trainee can form a decision on the readings and react accordingly. The simulated readings can be adjusted based on the reaction of the trainee and position of vehicle relative to hazard point.

A live virtual constructive (LVC) gateway system is configured to transparently separate, merge, and route data traffic between operator systems, live tactical Line Replaceable Unit (LRU) systems, and simulated tactical LRU systems. The LVC gateway is configured to receive LRU commands from an operator system, parse, the commands, and reconstruct the commands suitable for transmission to live or simulated tactical LRU systems. The LVC gateway is also configured to receive live and simulated status and target data from live tactical LRU and simulated tactical LRU systems, respectively, and merge the data for transmission to an operator system.

A live virtual constructive (LVC) gateway system is configured to transparently separate, merge, and route data traffic between operator systems, live tactical Line Replaceable Unit (LRU) systems, and simulated tactical LRU systems. The LVC gateway is configured to receive LRU commands from an operator system, parse, the commands, and reconstruct the commands suitable for transmission to live or simulated tactical LRU systems. The LVC gateway is also configured to receive live and simulated status and target data from live tactical LRU and simulated tactical LRU systems, respectively, and merge the data for transmission to an operator system.

A machine learning model is trained by defining a scenario including models of vehicles and a typical driving environment. A model of a subject vehicle is added to the scenario and sensor locations are defined on the subject vehicle. A perception of the scenario by sensors at the sensor locations is simulated. The scenario further includes a model of a parked vehicle with its engine running. The location of the parked vehicle and the simulated outputs of the sensors perceiving the scenario are input to a machine learning algorithm that trains a model to detect the location of the parked vehicle based on the sensor outputs. A vehicle controller then incorporates the machine learning model and estimates the presence and/or location of a parked vehicle with its engine running based on actual sensor outputs input to the machine learning model.

Embodiments of a system for simulating a radio frequency (RF) scene associated with a moving maritime surface are generally described herein. An RF scene is generated using an RF scene generation model and a moving maritime surface is generated using a maritime surface model. The RF scene is integrated with the moving maritime surface model. The RF scene generation model is configured to apply a radar model to generate and update the RF scene based on simulated radar returns at a radar pulse repetition frequency (PRF) and the maritime surface model is configured to update the moving maritime surface at a maritime surface update rate, access previous and current maritime surfaces, and interpolate surface facet properties to pulse times of the radar model, The maritime surface model is configured to update the moving maritime surface once every subdwell.