System, a method and a computer program product for maneuvering of an air vehicle
US-2015197335-A1 · Jul 16, 2015 · US
US9828107B1 · US · B1
| Field | Value |
|---|---|
| Publication number | US-9828107-B1 |
| Application number | US-201514835567-A |
| Country | US |
| Kind code | B1 |
| Filing date | Aug 25, 2015 |
| Priority date | Aug 25, 2014 |
| Publication date | Nov 28, 2017 |
| Grant date | Nov 28, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
The present invention provides a vehicle with redundant systems to increase the overall safety of the vehicle. In other aspects, the present invention provides a method for learning control of non-linear motion systems through combined learning of state value and action-value functions.
Opening claim text (preview).
What is claimed is: 1. A vertical take-off and landing (VTOL) vehicle, comprising: a frame having a plurality of opposingly located propulsion units attached thereto; each of said propulsion units comprising a plurality of microprocessors, a plurality of speed controllers, two motors that drive propellers, and a plurality of power sources that are electrically interconnected by at least one switch; said microprocessors monitor the operational state of said propulsion units and control said electrical interconnections; and each of said speed controllers coupled to one of said power sources and control the operation of said motors. 2. The VTOL vehicle of claim 1 , wherein said speed controllers have a three-phase output and inputs comprising a control signal received from one of said microprocessors and a sensed power level. 3. The VTOL vehicle of claim 1 , wherein said microprocessors include self-diagnostic capabilities. 4. The VTOL vehicle of claim 1 , wherein said microprocessors are trained for component failure and corresponding change in flight characteristics. 5. The VTOL vehicle of claim 1 , wherein said microprocessors are trained to maximize operational efficiency of said vehicle. 6. The VTOL vehicle of claim 1 , wherein said microprocessors are trained to maximize operational efficiency of said vehicle by alternating the use of the power sources that supply each motor to minimize temperature and maximize efficiency. 7. The VTOL vehicle of claim 1 , wherein in the event of a failure of one of said speed controllers, the motor assigned to the failed speed controller is operated by one of the remaining speed controllers. 8. The VTOL vehicle of claim 1 , wherein in the event of a failure of a power source assigned to a motor, one of said remaining power sources is used to power said motor. 9. The VTOL vehicle of claim 1 , wherein two microprocessors are provided and are arranged in a master/slave relationship. 10. The VTOL vehicle of claim 1 , wherein each propulsion unit performs decision-making based on internal monitoring of voltages, currents and power temperatures. 11. The VTOL vehicle of claim 1 , wherein each propulsion unit is adapted to autonomously select the power source coupled to a motor; to isolate a faulty power source; and under normal conditions, to switch between power sources coupled to a motor to minimize heat dissipation and maximize efficiency. 12. The VTOL vehicle of claim 1 , wherein each propulsion unit is adapted to autonomously control the distribution of power between said motors, and in the event one motor fails, it supplies power to another motor. 13. The VTOL vehicle of claim 1 , further including at least one artificial intelligent control unit and wherein each propulsion unit is adapted to constantly monitor and to report unusual component behavior before failure occurs to said at least one artificial intelligent control unit. 14. The VTOL vehicle of claim 13 , further including a second artificial intelligent control unit and wherein said artificial intelligent control units are arranged in a master/slave relationship. 15. The VTOL vehicle of claim 14 , wherein said artificial intelligent control units are adapted to produce motor controls for said propulsion units that take into account the current state-of-health of the vehicle. 16. The VTOL vehicle of claim 15 , wherein said artificial intelligent control units are adapted to produce motor controls for said propulsion units that take into account the current state-of-health of the vehicle by being trained under normal flying conditions to build a model that takes the measurements of actual flight data performance and compares said data to expected data from said model. 17. The VTOL vehicle of claim 16 , wherein when one of said artificial intelligent control units recognize a disruption from normal flight, said artificial intelligent control unit performs decision making to alter flight control, and depending on the disturbance pattern, said artificial intelligent control unit is adapted to perform an emergency landing or load balance said propulsion units and resume normal flight. 18. The VTOL vehicle of claim 16 , wherein said artificial intelligent control units are trained using reinforcement learning. 19. The VTOL vehicle of claim 18 , wherein said reinforcement learning comprises continuous action fitted value iteration. 20. The VTOL vehicle of claim 19 , wherein said continuous action fitted value iteration uses an algorithm comprising: Input: current state x, parametrization estimate θ Input: basis function vector F, simulator D Output: û 1: for i = 1, . . . , d u do 2: sample input U i = [u 1,i . . . u d α ,i ] T 3: for j = 1, . . . , d n do 4: x′ j,i ← D(x, u j,i ) 5: Q x,j (u j,i ) ← θ T F(x′ j,i ) 6: end for 7: {circumflex over (p)} i ← argmin pi Σ j=1 d n (C j,i p i − Q x,j (u j,i )) 2 8: u ^ i * ← - p 1 , i 2 p 2 , i 9: û i = min(max(û i *, u i I ), u i u ) 10: end for 11: û ← calculated with (9) 12: return û.
Probabilistic graphical models, e.g. probabilistic networks · CPC title
using redundant signals or controls · CPC title
Control of attitude, i.e. control of roll, pitch, or yaw · CPC title
Rotorcraft; Rotors peculiar thereto · CPC title
Arrangement of jet reaction apparatus for propulsion or directional control · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.