Intelligent autonomous and semi-autonomous variable altitude vehicle deployment system and method

1 . A computer-implemented method of providing services to a disaster-impacted location, the method comprising:

automatically determining, using a damage detection module executing on a controller comprising a processor, that a first type of damage has impacted a first location;

automatically generating, using a damage classifier executing on the controller at the damage detection module, a first damage assessment including details regarding the first type of damage;

communicating the first damage assessment to an autonomous vehicle transport manager executing on the controller;

using the autonomous vehicle transport manager to identify a plurality of autonomous vehicles from a fleet of autonomous vehicles based on the first damage assessment, each autonomous vehicle in the plurality of autonomous vehicles performing a specific task or set of tasks;

loading the plurality of autonomous vehicles on a transport vehicle;

moving the transport vehicle to a position near the first location;

receiving, at a deployment manager executing on the controller and from the damage detection module, the first damage assessment;

automatically selecting, using the first damage assessment as an input in a machine learning model running in an autonomous vehicle assignment generator on the deployment manager a mixed pattern and number of one or more autonomous vehicles from the plurality of autonomous vehicles on the transport vehicle likely to fulfill requirements and perform needed response tasks in the first location to include in an autonomous vehicle party for deployment to the first location;

transmitting schedules and assignments for the selected one or more autonomous vehicles in the autonomous vehicle party from the autonomous vehicle assignment generator to the autonomous vehicle transport manager, the autonomous vehicle transport manager providing instructions to the selected one or more autonomous vehicles for deployment to the first location; and

deploying the selected one or more autonomous vehicles from the transport vehicle to the first location.

2 . The method of claim 1 , further comprising:

receiving, at the damage detection module, a plurality of aerial sensor data of the first location; and

employing a deep learning damage classification model using the plurality of aerial sensor data to identify the first type of damage.

3 . The method of claim 1 , wherein the fleet of autonomous vehicles is selected based on the fleet of autonomous being nearest to the first location.

4 . The method of claim 1 , further comprising receiving, at the deployment manager, on-site sensor information collected by vehicles in the first location.

5 . The method of claim 1 , further comprising determining, at the deployment manager, that the plurality of autonomous vehicles on the transport vehicle is at a first distance from the first location, wherein the one or more autonomous vehicles are selected from a subset of autonomous vehicles on the transport vehicle with a battery life sufficient to permit travel across at least the first distance, perform a given specific task or set of tasks, and return to the transport vehicle.

6 . The method of claim 1 , wherein the first type of damage is one of flood, fire, tornado, landslide, wind, and hail.

7 . The method of claim 6 , wherein the one or more autonomous vehicles are selected from a subset of autonomous vehicles on the transport vehicle with onboard sensors capable of collecting data while exposed to conditions associated with the first type of damage.

8 . The method of claim 2 , further comprising identifying, using the deep learning damage classification model, a damage magnitude rating associated with the first location, wherein a number of autonomous vehicles selected for inclusion in the autonomous vehicle party is based on the identified damage magnitude rating.

9 . The method of claim 8 , wherein the number of autonomous vehicles that are selected when the identified damage magnitude rating is above a preselected level is greater than when the identified damage magnitude rating is below the preselected level.

10 . A damage response system for providing services to a disaster-impacted location, the system comprising:

a damage detection module executing on a controller comprising a processor to generate a damage assessment based on a plurality of aerial sensor data of a first location;

an autonomous vehicle transport manager executing on the controller to identify a plurality of autonomous vehicles from a fleet of autonomous vehicles based on the damage assessment, each autonomous vehicle in the plurality of autonomous vehicles performing a specific task or set of tasks; and

a deployment manager executing on the controller to use the first damage assessment as an input in a machine learning model running in an autonomous vehicle assignment generator on the deployment manager to select an autonomous vehicle party comprising a mixed pattern and number of one or more autonomous vehicles from a plurality of autonomous vehicles selected from the fleet of autonomous vehicles loaded onto a transport and transported to a position near the first location for deployment to the first location based on the damage assessment that are likely to fulfill requirements and perform needed response tasks in the first location.

11 . The damage response system of claim 10 , wherein the controller triggers deployment of the autonomous vehicle party to the first location in response to instructions received from the deployment manager.

12 . The damage response system of claim 10 , wherein the plurality of autonomous vehicles is loaded on the transport vehicle by an autonomous vehicle transport manager executing on the controller.

13 . A system for providing services to a disaster-impacted location, the system comprising a processor and machine-readable media including instructions which, when executed by the processor, cause the processor to:

automatically determine, using a damage detection module executing on a controller comprising a processor, that a first type of damage has impacted a first location;

automatically generate, using a damage classifier executing on the controller, a first damage assessment including details regarding the first type of damage;

communicate the first damage assessment to an autonomous vehicle transport manager executing on the controller;

use the autonomous vehicle transport manager to identify a plurality of autonomous vehicles from a fleet of autonomous vehicles based on the first damage assessment, each autonomous vehicle in the plurality of autonomous vehicles performing a specific task or set of tasks;

load the plurality of autonomous vehicles on a transport vehicle;

move the transport vehicle to a position near the first location;

receive, at a deployment manager executing on the controller and from the damage detection module, the first damage assessment;

automatically select, using the first damage assessment as an input in a machine learning model running in an autonomous vehicle assignment generator on the deployment manager a mixed pattern and number of one or more autonomous vehicles from the plurality of autonomous vehicles on the transport vehicle likely to fulfill requirements and perform needed response tasks in the first location to include in an autonomous vehicle party for deployment to the first location;

transmit schedules and assignments for the selected one or more autonomous vehicles in the autonomous vehicle party from the autonomous vehicle assignment generator to the autonomous vehicle transport manager, the autonomous vehicle transport manager providing instructions to the selected one or more autonomous vehicles for deployment to the first location; and

deploy the selected one or more autonomous vehicles from the transport vehicle to the first location.

14 . The system of claim 13 , wherein the instructions further cause the processor to:

receive, at the damage detection module, a plurality of aerial sensor data of the first location; and

employ a deep learning damage classification model using the plurality of aerial sensor data to identify the first type of damage.

15 . The system of claim 13 , wherein the fleet of autonomous vehicles is selected based on the fleet of autonomous vehicles being nearest to the first location.

16 . The system of claim 13 , wherein the instructions further cause the processor to receive, at the deployment manager, on-site sensor information collected by vehicles in the first location.

17 . The system of claim 13 , wherein the instructions further cause the processor to determine, at the deployment manager, that the plurality of autonomous vehicles on the transport vehicle is at a first distance from the first location, wherein the one or more autonomous vehicles are selected from a subset of autonomous vehicles on the transport vehicle with a battery life sufficient to permit travel across at least the first distance, perform a given specific task or set of tasks, and return to the transport vehicle.

18 . The system of claim 13 , wherein the first type of damage is one of flooding, fire, wind, and hail.

19 . The system of claim 18 , wherein the one or more autonomous vehicles are selected from a subset of autonomous vehicles on the transport vehicle with onboard sensors capable of collecting data while exposed to conditions associated with the first type of damage.

20 . The system of claim 14 , wherein the instructions further cause the processor to identify, using the deep learning damage classification model, a damage magnitude rating associated with the first location, wherein a number of autonomous vehicles selected for inclusion in the autonomous vehicle party is based on the identified damage magnitude rating.