A device includes a control input generator and a neural network trainer. A flight simulator is configured to generate first state data responsive to a first control input from the control input generator and to provide the first state data to a first neural network to generate a candidate second control input. The control input generator is also configured to select, based on a random value, a second control input from between the candidate second control input and a randomized offset control input that is based on a random offset applied to the first control input. The flight simulator is configured to generate second state data responsive to the second control input from the control input generator. The neural network trainer is configured to update weights of the first neural network based, at least in part, on the first state data and the second state data.

A device includes a control input generator and a neural network trainer. A flight simulator is configured to generate first state data responsive to a first control input from the control input generator and to provide the first state data to a first neural network to generate a candidate second control input. The control input generator is also configured to select, based on a random value, a second control input from between the candidate second control input and a randomized offset control input that is based on a random offset applied to the first control input. The flight simulator is configured to generate second state data responsive to the second control input from the control input generator. The neural network trainer is configured to update weights of the first neural network based, at least in part, on the first state data and the second state data.

In one aspect, a system for people to simulate flying motion may include a plurality of supporting posts, one or more control units, and each of the control unit is connected to the user's body through a string, and a sensor is disposed on each string. The sensor can detect the tension change of the string and transmit the signal to the control unit to adjust the tension accordingly. When the user moves the wings, the change of the tension can be detected by the sensor and transmitted to the control unit to further adjust the tension of each string by pulling or releasing the strings to enable the user to enjoy the flying motion.

In one aspect, a system for people to simulate flying motion may include a plurality of supporting posts, one or more control units, and each of the control unit is connected to the user's body through a string, and a sensor is disposed on each string. The sensor can detect the tension change of the string and transmit the signal to the control unit to adjust the tension accordingly. When the user moves the wings, the change of the tension can be detected by the sensor and transmitted to the control unit to further adjust the tension of each string by pulling or releasing the strings to enable the user to enjoy the flying motion.

A method of operating a targeting system simulation tool (TSST) includes providing a TSST configured to receive an obstacle/effect parameterization, a simulation parameterization, a sensor parameterization, an aircraft parameterization, and an autonomy parameterization. The method further includes receiving by the TSST, at least one of each of an obstacle/effect parameterization and a simulation parameterization. The method further includes receiving by the TSST, either (1) a sensor parameterization or (2) an aircraft parameterization. The method further includes operating the TSST to apply the provided ones of the obstacle/effect parameterization, simulation parameterization, sensor parameterization, and aircraft parameterization to generate a value, value range, or value limit for the unprovided aircraft parameterization or unprovided sensor parameterization.

A device includes a control input generator and a neural network trainer. A flight simulator is configured to generate first state data responsive to a first control input from the control input generator and to provide the first state data to a first neural network to generate a candidate second control input. The control input generator is also configured to select, based on a random value, a second control input from between the candidate second control input and a randomized offset control input that is based on a random offset applied to the first control input. The flight simulator is configured to generate second state data responsive to the second control input from the control input generator. The neural network trainer is configured to update weights of the first neural network based, at least in part, on the first state data and the second state data.

A device includes a control input generator and a neural network trainer. A flight simulator is configured to generate first state data responsive to a first control input from the control input generator and to provide the first state data to a first neural network to generate a candidate second control input. The control input generator is also configured to select, based on a random value, a second control input from between the candidate second control input and a randomized offset control input that is based on a random offset applied to the first control input. The flight simulator is configured to generate second state data responsive to the second control input from the control input generator. The neural network trainer is configured to update weights of the first neural network based, at least in part, on the first state data and the second state data.

A vehicle nonlinear dynamics experimental simulation device, such as flight simulator, including a motorized spherical vehicle suspended inside a spherical shell which has a smooth inner surface. The spherical vehicle is supported by a plurality of spiky legs with bearing assemblies. The spherical shell is supported by three controllable translational motion platforms. Simulating apparatuses for a pilot cabin is mounted inside the spherical vehicle. The spherical vehicle has driving, restoring, and damping capabilities in roll, pitch, and yaw directions and is capable of unbounded rotation in any directions. The spherical vehicle provides an experimental model to simulate a vehicle's rotational dynamics.

A vehicle nonlinear dynamics experimental simulation device, such as flight simulator, including a motorized spherical vehicle suspended inside a spherical shell which has a smooth inner surface. The spherical vehicle is supported by a plurality of spiky legs with bearing assemblies. The spherical shell is supported by three controllable translational motion platforms. Simulating apparatuses for a pilot cabin is mounted inside the spherical vehicle. The spherical vehicle has driving, restoring, and damping capabilities in roll, pitch, and yaw directions and is capable of unbounded rotation in any directions. The spherical vehicle provides an experimental model to simulate a vehicle's rotational dynamics.

The invention relates to a method for operating a driving simulator having the following steps: detecting a braking request in the driving simulator, in particular on the basis of actuation of a brake actuator; converting the detected braking request into at least one braking signal suitable for characterising the braking request; transmitting the at least one braking signal from the driving stimulator to a test bench on which are mounted at least part of a drivetrain with at least one axle section of a vehicle, in particular an axle half, and at least one brake associated with the at least one axle section; rotating the at least one axle section at a wheel speed which corresponds to a predefined speed of the vehicle; actuating the at least one brake of the vehicle on the basis of the at least one braking signal; setting a predefined torque or a predefined wheel speed of at least one axle section of the at least one dynamometer on the basis of properties of at least one component of the vehicle, in particular of the drivetrain, of the vehicle and/or of the entire vehicle, wherein the properties are at least partially simulated; detecting the actual wheel speed at a predefined torque or the actual torque at a predefined wheel speed; and outputting the actual wheel speed or the actual torque to the driving simulator.