Systems and methods for real-time, multi-factor prediction of emergency landing success probabilities using machine learning
Embodiments of the present disclosure provide systems and methods for real-time, multi-factor prediction of emergency landing success probabilities using machine learning. In one embodiment, a method includes generating, by one or more processors, a first emergency landing success probability value, the first emergency landing success probability value based at least in part on one or more vehicle state conditions for a vehicle; generating, by the one or more processors, a second emergency landing success probability value, the second emergency landing success probability value based at least in part on one or more environmental state conditions for an operating environment of the vehicle; and generating, by the one or more processors, a third emergency landing success probability value based at least in part on the first emergency landing success probability value and the second emergency landing success probability value.
1 . A method comprising:
retrieving a vehicle state condition of a vehicle;
applying the vehicle state condition to a vehicle state machine learning model to generate a vehicle state related emergency landing success probability value indicative of a first probability that the vehicle will perform a successful emergency landing procedure;
retrieving a communication link condition of the vehicle;
applying the communication link condition to a communication link machine learning model to generate a communication link related emergency landing success probability value indicative of a second probability that the vehicle will perform the successful emergency landing procedure, wherein the first probability is different than the second probability;
applying the vehicle state related emergency landing success probability value and the communication link related emergency landing success probability value to an emergency landing machine learning model to generate an emergency landing success probability value indicative of a third probability value that the vehicle will perform the successful emergency landing procedure; and
causing initiation of an emergency landing procedure based on the emergency landing success probability value.
2 . The method of claim 1 , wherein:
the vehicle state machine learning model is trained using (i) a first training dataset generated using one or more first simulations and (ii) a second training dataset comprising first historical vehicle incident data; and
the communication link machine learning model is trained using (a) a third training dataset generated using one or more second simulations and (b) a fourth training dataset comprising second historical vehicle incident data.
3 . The method of claim 1 , further comprising:
generating a landing port related emergency landing success probability value, the landing port related emergency landing success probability value based at least in part on one or more landing port conditions, wherein the emergency landing success probability value is based at least in part on the landing port related emergency landing success probability value.
4 . The method of claim 3 , further comprising:
providing, to a ground-based computing device, a recommendation for improving the landing port related emergency landing success probability value, wherein the recommendation comprises an indication to clear a potential emergency landing site of one or more obstacles.
5 . The method of claim 1 , further comprising:
providing, to a user interface of the vehicle, a recommendation for improving the vehicle state related emergency landing success probability value.
6 . The method of claim 1 , wherein one or more of (i) the vehicle state related emergency landing success probability value, (ii) the communication link related emergency landing success probability value, or (iii) the emergency landing success probability value are generated periodically during a flight time period for the vehicle.
7 . The method of claim 1 , wherein generating the emergency landing success probability value is triggered by one or more of (i) the vehicle state related emergency landing success probability value satisfying a first threshold or (ii) the communication link related emergency landing success probability value satisfying a second threshold.
8 . A system comprising:
a user interface; and
one or more processors in communication with the user interface, the one or more processors configured to:
retrieve a vehicle state condition of a vehicle;
apply the vehicle state condition to a vehicle state machine learning model to generate a vehicle state related emergency landing success probability value indicative of a first probability that the vehicle will perform a successful emergency landing procedure;
retrieve a communication link condition of the vehicle;
apply the communication link condition to a communication link machine learning model to generate a communication link related emergency landing success probability value indicative of a second probability that the vehicle will perform the successful emergency landing procedure, wherein the first probability is different than the second probability;
apply the vehicle state related emergency landing success probability value and the communication link related emergency landing success probability value to an emergency landing machine learning model to generate an emergency landing success probability value indicative of a third probability value that the vehicle will perform the successful emergency landing procedure; and
cause initiation of an emergency landing procedure based on the emergency landing success probability value.
9 . The system of claim 8 , wherein:
the vehicle state machine learning model is trained using (i) a first training dataset generated using one or more first simulations and (ii) a second training dataset comprising first historical vehicle incident data; and
the communication link machine learning model is trained using (a) a third training dataset generated using one or more second simulations and (b) a fourth training dataset comprising second historical vehicle incident data.
10 . The system of claim 8 , wherein the one or more processors are further configured to:
generate a landing port related emergency landing success probability value, the landing port related emergency landing success probability value based at least in part on one or more landing port conditions, wherein the emergency landing success probability value is based at least in part on the landing port related emergency landing success probability value.
11 . The system of claim 10 , wherein the one or more processors are further configured to:
provide, to a ground-based computing device, a recommendation for improving the landing port related emergency landing success probability value, wherein the recommendation comprises an indication to clear a potential emergency landing site of one or more obstacles.
12 . The system of claim 8 , wherein the one or more processors are further configured to:
provide, to the user interface, a recommendation for improving the emergency landing success probability value.
13 . The system of claim 8 , wherein one or more of (i) the vehicle state related emergency landing success probability value, (ii) the communication link related emergency landing success probability value, or (iii) the emergency landing success probability value are generated periodically during a flight time period for the vehicle.
14 . The system of claim 8 , wherein generating the emergency landing success probability value is triggered by one or more of (i) the vehicle state related emergency landing success probability value satisfying a first threshold or (ii) the communication link related emergency landing success probability value satisfying a second threshold.
15 . An apparatus comprising:
one or more processors; and
a memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
retrieve a vehicle state condition of a vehicle;
apply the vehicle state condition to a vehicle state machine learning model to generate a vehicle state related emergency landing success probability value indicative of a first probability that the vehicle will perform a successful emergency landing procedure;
retrieve a communication link condition of the vehicle;
apply the communication link condition to a communication link machine learning model to generate a communication link related emergency landing success probability value indicative of a second probability that the vehicle will perform the successful emergency landing procedure, wherein the first probability is different than the second probability;
apply the vehicle state related emergency landing success probability value and the communication link related emergency landing success probability value to an emergency landing machine learning model to generate an emergency landing success probability value indicative of a third probability value that the vehicle will perform the successful emergency landing procedure; and
cause initiation of an emergency landing procedure based on the emergency landing success probability value.
16 . The apparatus of claim 15 , wherein:
the vehicle state machine learning model is trained using (i) a first training dataset generated using one or more first simulations and (ii) a second training dataset comprising first historical vehicle incident data; and
the communication link machine learning model is trained using (a) a third training dataset generated using one or more second simulations and (b) a fourth training dataset comprising second historical vehicle incident data.