Rockwell Collins, Inc.

Joshua R Bertram

Angus L McLean

Brian R Wolford

Alexander Roup

James A Marek

Jaclyn A Hoke

Jason A Myren

Branden H Sletteland

Roger K Shultz

Scott S Rediger

Raymond Knoff

Reginald D Bean

Max G Taylor

John L Hagen

Thomas L Vogl

Levi D Van Oort

Maxime M Gariel

Lloyd F Aquino

David W Hubin

Ian McElhenny

Matthew A Griess

Andre F Mitchell

Curtis M Topf

Wade D Paustian

A method for testing platforms (e.g., live, virtual, and/or constructive platforms associated with autonomous aircraft systems and their component subsystems) in a live/virtual/constructive (LVC) environment. In embodiments, the method includes determining, via a testbed engine, the development state of a platform component under test. The method includes retrieving a test to be executed, the test including test conditions to be applied to the component. The method includes determining whether the component is enabled to respond to the test conditions. The method includes, if the component is enabled to respond to the test conditions, executing the test while monitoring the component to detect a first output response and a second output response. The method includes identifying, via the testbed engine, at least one change in the development state of the component by comparing the first and second output responses.

Systems and methods for autonomous vehicle systems testing

A system includes a machine learning engine configured to receive training data including a plurality of input conditions associated with a state space and a plurality of response maneuvers associated with the state space and train a learning system using the training data and a reward function including a plurality of terms associated with a plurality of end state spaces, each term in the plurality of terms defines an end reward value for each end state space. A value function and policy are generated. The value function comprising a plurality of values, wherein each response maneuvers in the plurality of response maneuvers is associated with a value in the plurality of values related to transitioning from the state space to each end state space, the policy indicative of connections between the state spaces, plurality of values, and the respective end reward value for the plurality of end state spaces.

Systems and methods for practical autonomy decision controller

A system for manufacturing, testing, integrating, and operating of live or virtual platforms and components thereof uses a simulation engine and/or a testbed engine in a live-virtual-constructive (LVC) environment. The system can be used in a pure simulated environment or for testing of individual subsystems on a live platform (e.g., a live aircraft) with remaining subsystems in simulation, to incremental integration of all subsystems onto live aircraft.

Machine learning, evaluating, and reinforced learning within systems or apparatuses enables autonomy to a complexity level beyond automation. Inferences are made using machine learning based on observations, images, or video feed of operator input. The inferences are evaluated or classified and maneuvers are performed based on the evaluating or the classification. The performed maneuvers may be further evaluated for scoring or weighting. The reinforcement learning may perform updates based on the scoring, weighting, and a maximizing reward function such that the machine learning is constantly improving.

Learning from operator data for practical autonomy

Safe practical autonomy is ensured by encapsulating an unreliable or untrusted machine learning algorithm within a control-based algorithm. A safety envelope is utilized to ensure that the machine learning algorithm does not output control signals that are beyond safe thresholds or limits. Secure practical autonomy is ensured by verification using digital certificates or cryptographic signatures. The verification may be for individual partitions of an autonomous system or apparatus. The partitions include trusted and untrusted partitions. Trusted partitions are verified for security, while untrusted partitions are verified for safety and security.

Safe and secure practical autonomy

Systems for mobile launch and retrieval of unmanned aircraft systems (UAS) include a rail-based, ground-based, or water-based mobile platform carrying a mobile station for launching and retrieving vertical take-off and landing (VTOL) or non-VTOL UAS while the platform is in motion, based on current position and weather conditions. The mobile platform may include facilities for communicating with the airborne UAS, stowing a retrieved UAS, and reloading/refitting a stowed UAS. The mobile platform may include positionable wake control devices and a partially positionable launch and retrieval mechanism for alleviating turbulence or crosswinds. The mobile platform may include long-range sensors for detecting or identifying obstacles near a rail-based platform that may interfere with the operating envelope of the launch and retrieval mechanisms. The launch and retrieval system may be intermodal and scalable either up or down as mission parameters demand.

Mobile UAV launch and retrieval system for operation within a transportation system

A relative positioning system is described. At least one emitter is attached to a first object, where each of the at least one emitters includes: an electromagnetic radiation source configured to generate electromagnetic radiation over a band of wavelengths, and a prism arranged to refract and disperse the electromagnetic radiation from the electromagnetic radiation source according to the wavelength of the electromagnetic radiation. At least one electromagnetic radiation detector is attached to a second object arranged to detect the wavelengths of some of the electromagnetic radiation refracted and dispersed by a respective prism. At least one processor is configured to determine the relative position of the first object and the second object based on the detected wavelengths by the at least one electromagnetic radiation detector.

Prismatic relative positioning system

A system and related method for PCIe device configuration in a certified multi-core avionics processing system on which several guest operating systems (GOS) are running may allow a GOS to access or communicate with PCIe devices not owned by that GOS. The system may configure PCIe controllers and the PCI devices connected by those controllers by issuing addresses and determine, via a configuration vector of the system hypervisor, which PCIe devices are accessible to which non-owning guest operating systems. The hypervisor may provide each non-owning GOS with the GOS physical addresses corresponding to each non-owned PCIe device accessible to that GOS. Configuration of an unpowered or otherwise unprepared PCIe device may be deferred until device information is requested by the owning GOS to preserve compliance with system timing requirements.

Multicore safe PCIe solution

An avionics development environment based on high level interpreted language for rapid creation and deployment of avionics software is disclosed. Functional modules are segregated in time and allocated segregated resources so that functional modules only interact in predictable, deterministic ways. Segregated functional modules are individually certifiable for avionics operation, and parameters necessary for certification are associated with each functional module to ensure the end application conforms to such parameters.

Rapid avionics development environment

A system includes a machine learning engine. The machine learning engine is configured to receive training data including a plurality of first input conditions and a plurality of first response maneuvers associated with the first input conditions. The machine learning engine is configured to train a learning system using the training data to generate a second response maneuver based on a second input condition.

Systems and methods for machine learning of pilot behavior

A multicore adaptive scheduler of tasks in an ARINC 653-compliant avionics system allocates flight critical tasks execution time equivalent to their worst case execution time and allocates quality-driven tasks minimum execution time equivalent to their minimum completion time. The scheduler may also offset the start time of a task or define an upper bound for completion time of a quality-driven task. The scheduler generates and executes partition schedules of tasks, reallocating execution time unused by completed tasks and reallocating execution time from interrupt handlers to tasks preempted by interrupts. The scheduler may also analyze the viability of a generated schedule. The scheduler uses rate limiting and flow control techniques to ensure a predictable amount of execution time to be reallocated for interrupt handling.

Multicore adaptive scheduler

A system, device, and method for generating and employing risk-based flight path data are disclosed. The system for employing risk-based flight path data may include one or more one avionics systems and/or remote aircraft operator systems configured to receive risk-based flight path data from a route generator (RG). The RG may acquire navigation data representative of one or more waypoints, acquire risk object data based upon the navigation data, determine the risk-based flight path data representative of a risk-based flight path as a function of the acquired navigation data, the acquired risk data, and a route generating algorithm, and provide the flight path data to the one or more avionics systems and/or remote aircraft operator systems. In some embodiments, the risk object data may include a plurality of risk clearance altitudes. In other embodiments, the risk object data may include a plurality of risk clearance elevations.

Risk-based flight path data generating system, device, and method

Systems and methods for providing dynamic libraries in safety critical computing environments are disclosed. Controlled dynamic libraries and isolated execution spaces are utilized. In some embodiments, the controlled dynamic libraries and isolated execution spaces are implemented in full compliance with rules and standards established by aviation regulatory and government agencies, allowing systems utilizing the controlled dynamic libraries to be certifiable for avionics.

Computing systems utilizing controlled dynamic libraries and isolated execution spaces

A method and system for securely distributing human-machine input/output to multi-level displays in a multi-level security environment is disclosed. The method and system in accordance with the present disclosure provides the ability to take input from common input devices and manages the input to ensure that the input is delivered only to the intended security domain/level and that the input is delivered only to the intended display element within the intended security domain/level. Furthermore, architectures configured for supporting the multi-level security display with secure input/output are also disclosed.

Multi-level security display with secure input/output

A secure deterministic fabric includes switches that segregate data traffic requiring disparate levels of authentication or having different safety levels. Data may be segregated physically, utilizing different hardware; or virtually, by allocating certain assets such as memory blocks exclusively for certain levels of authentication. The secure deterministic fabric may include elements for safety monitoring and multi-level security monitoring.

Growing community of inventors

Ames, IA, United States of America

Joshua R Bertram