Abstract: A method for agent and scenario modeling is described. The method includes analyzing, through a model-based systems engineering (MBSE) model, high-level events extracted from driving log data to identify a dataset of interest from the driving log data. The method also includes extracting an episode of interest from the dataset of interest according to an MBSE state transition diagram. The method further includes generating a driving scenario of interest based on the episode of interest. The method also includes utilizing the driving scenario of interest to parameterize agent and/or scenario models used for autonomous operation of an ego vehicle.

Abstract: A simulation system includes a storage module comprising a plurality of behaviors and an electronic control unit. The electronic control unit is configured to define an agent for use in a simulation, where the agent comprises one or more of the plurality of behaviors from the storage module and the agent is capable of implementation on a plurality of different simulators, establish a contract with a first simulator and the agent to provide a consistent boundary between the agent and the first simulator, and execute the simulation of a device in a simulation environment with the first simulator and the agent.

Abstract: Systems and methods for scenario marker infrastructure are disclosed. In one embodiment, a method comprises receiving driving scenario data, extracting simulated sensor data from the driving scenario data, extracting scenario markers from the driving scenario data, inputting the simulated sensor data into autonomous driving software, receiving driving instructions output by the driving software, updating a driving status of the autonomous vehicle based on the driving instructions output by the autonomous driving software, creating a simulation log comprising the simulated sensor data and the driving instructions, and inserting the scenario markers into the simulation log.

Abstract: A simulation system includes a storage module comprising a plurality of behaviors and an electronic control unit. The electronic control unit is configured to define an agent for use in a simulation, where the agent comprises one or more of the plurality of behaviors from the storage module and the agent is capable of implementation on a plurality of different simulators, establish a contract with a first simulator and the agent to provide a consistent boundary between the agent and the first simulator, and execute the simulation of a device in a simulation environment with the first simulator and the agent.

Abstract: Systems and methods for scenario marker infrastructure are disclosed. In one embodiment, a method comprises receiving driving scenario data, extracting simulated sensor data from the driving scenario data, extracting scenario markers from the driving scenario data, inputting the simulated sensor data into autonomous driving software, receiving driving instructions output by the driving software, updating a driving status of the autonomous vehicle based on the driving instructions output by the autonomous driving software, creating a simulation log comprising the simulated sensor data and the driving instructions, and inserting the scenario markers into the simulation log.

Abstract: Systems and methods are provided for validating a map via automatic test generation for multiple test route subsections. Autonomous vehicles required accurate maps. As such, new maps may be added and/or maps map be updated frequently. Verifying these maps requires testing. Maps may be validated via automatic test generation by obtaining route information identifying a route in the map for an autonomous vehicle. The route may be segmented into multiple test route subsections. Features corresponding to one or more of the test route subsections may be identified. One or more validation tests corresponding to the test route subsections may be generated based on a simulated autonomous vehicle traversing the one or more road features. For example, multiple validation tests for corresponding test route subsections may be generated and executed in parallel.