Neurocomputing control distribution system

A control system for controlling a vehicle, said vehicle containing a control subsystem including at least one propulsion, aerodynamic, or other control effector, said control system comprising:

• (a) at least one sensor for sensing vehicle position and motions and operating conditions;
• (b) control means for generating control input signals;
• (c) a generator for generating desired vehicle forces/moments from the sensed vehicle position and motions and the generated control input signals based on predefined vehicle compensation and control laws;
• and
• (d) a neural network controller for generating commands for the at least one control effector by propagating the generated desired vehicle forces/moments, and operating conditions, through the neural network controller.

A method for controlling the movement of a thrust-propelled vehicle, comprising:

• (a) training a neural network controller off-line, said training comprising:
• (i) generating a plurality of possible system commands based on possible conditions of a thrust-propelled vehicle;
• (ii) establishing a neural network structure;
• and
• (iii) generating nodal weight values for the established neural network structure for each of the possible system commands;
• (b) sensing a condition of the thrust-propelled vehicle;
• and
• (c) the neural network controller generating at least one system command based on the sensed condition of the thrust-propelled vehicle and the nodal weight value generated off-line for the condition of the thrust-propelled vehicle.

A computer-readable medium having computer-executable instructions for performing a method for controlling the movement of a thrust-controlled vehicle, comprising:

• (a) training a neural network controller off-line, said training comprising:
• (i) generating a plurality of possible system commands based on possible conditions of a thrust-propelled vehicle;
• (ii) establishing a neural network structure;
• and
• (iii) generating nodal weight values for the neural network controller for each of the possible system commands;
• (b) sensing a condition of the thrust-propelled vehicle;
• and
• (c) the neural network controller generating at least one system command based on the sensed condition of the thrust-controlled vehicle and the nodal weight value generated off-line for the system command.

A method for controlling a vehicle, said vehicle containing a control subsystem including at least one propulsion, aerodynamic, or other control effector, said method comprising:

• (a) sensing vehicle position and motions and operating conditions;
• (b) generating control signals based on control inputs;
• (c) generating desired vehicle forces/moments from the sensed vehicle position and motions and the generated control input signals based on predefined vehicle compensation and control laws;
• and
• (d) generating commands for the at least one control effector by propagating the generated desired vehicle forces/moments and operating conditions through a neural network.

A computer-readable medium having computer-executable instructions for performing the method comprising:

• (a) sensing vehicle position and motions and operating conditions;
• (b) generating control signals based on control inputs;
• (c) generating desired vehicle forces/moments from the sensed vehicle position and motions and the generated control input signals based on predefined vehicle compensation and control laws;
• and
• (d) generating commands for the at least one control effector by propagating the generated desired vehicle forces/moments and operating conditions through a neural network.

A method for training a neural network to control a control subsystem of a vehicle, said control subsystem including at least one propulsion, aerodynamic, or other control effector, said method comprising:

• (a) generating sets of desired vehicle forces/moments for possible vehicle command input signals, and vehicle position and motions based on predefined vehicle compensation and control laws;
• (b) generating sets of control effector commands that yield feasible control subsystem forces/moments that correspond to sets of generated desired vehicle forces/moments;
• (c) establishing a neural network structure;
• and
• (d) generating nodal weight values for the established neural network structure based on the generated sets of generated desired vehicle forces/moments and the corresponding sets of control effector commands that yield feasible control subsystem forces/moments.

A computer-readable medium having computer-executable instructions for performing the method comprising:

• (a) generating sets of desired vehicle forces/moments for possible vehicle command input signals, and vehicle position and motions based on predefined vehicle compensation and control laws;
• (b) generating sets of control effector commands that yield feasible control subsystem forces/moments that correspond to sets of generated desired vehicle forces/moments;
• (c) establishing a neural network structure;
• and
• (d) generating nodal weight values for the established neural network structure based on the generated sets of generated desired vehicle forces/moments and the corresponding sets of control effector commands that yield feasible control subsystem forces/moments.