System and Method for Predicting An Adequate Ratio of Unmanned Vehicles to Operators

Patent metadata

Field	Value
Publication number	US-2016252902-A1
Application number	US-201615058917-A
Country	US
Kind code	A1
Filing date	Mar 2, 2016
Priority date	Nov 8, 2011
Publication date	Sep 1, 2016
Grant date	—

Abstract

The present invention is a computer decision tool for use in a system for controlling a team of unmanned vehicles. The computer decision tool includes a system performance model for receiving interface usability, automation level and algorithm efficiency variables and an operator performance model. The operator performance model receives task management efficiency and decision making strategy or DM efficiency variables. The system performance model is responsive to the interface usability, automation level and algorithm efficiency variables for providing a system performance status signal. The operator performance model is responsive to task management efficiency and DM strategy variables for providing an operator performance status signal. An operator capacity decision model is responsive to the system performance and operator performance status signals and a workload variable for providing a decision signal representative of an adequate team size or an optimal recommendation, such as changing the team size.

First claim

What is claimed is:

1 . A method for predicting the cognitive ability of an operator to control a plurality of unmanned vehicles, the method incorporating a system performance mode and an operator performance model and further comprising the steps of:
A) calculating a conditional probability for an interface usability decision node; B) using the results of said step A) to calculate an automation level decision node conditional probability; C) using the results of said step B) to calculate an algorithm efficiency decision node conditional probability; D) using the results of said step C) to calculate a task management efficiency decision node; E) using the results of said step D) to calculate an operator decision making efficiency decision nodes; F) calculating a conditional probability that said operator has sufficient said cognitive ability to control said plurality of unmanned vehicles using the results of said step E); the system performance model responsive to the usability, automation level and algorithm efficiency variables for providing a system performance status signal; and, the operator performance model responsive to the task management efficiency and operator decision making efficiency variables for providing an operator performance status signal

2 . The method of claim 1 wherein said step A) are calculated using conditional probability tables (CPTs) of nature and utility nodes, and wherein said CPT's are repopulated after the accomplishment of each of said steps A) through E).

3 . The method of claim 2 wherein said utility nodes include one or more utility node indications from the group consisting of Utility Interface; Utility Automation; Utility System DM; Utility Task Management; Utility Operator DM; and Utility of Capacity utility nodes.

4 . The method of claim 3 wherein said observable nature nodes include one or more observable nature node indications from the group consisting of Wait Times due to Loss of Situation Awareness (WTSA); Neglect Time (NT); Interaction Time (IT); Wait Times due to Queue (WTQ); Utilization Time (UT); System Interruption; Elimination Task Success; UV Health Status; Total Time for Target Elimination; Team Size; Team Heterogeneity; Frequency of Reassignment; Total Task Time; and ID Success Rate nodes.

5 . The method of claim 4 wherein said unobservable nature nodes which include one or more unobservable nature node indications from the group consisting of System Performance; Information Overload; Situation Awareness (SA); Workload; Wait Times due to Cognitive Reorientation; Automation Level; Operator Performance; Task Complexity; Enemy Team; and Task Heterogeneity unobservable nature nodes.

6 . In a system for controlling a team of unmanned vehicles, a computer decision tool comprising:
a system performance model responsive to an interface usability decision node, an adequate automation decision node, and an algorithm efficiency decision node; an operator performance model responsive to a task management efficiency decision node and an operator decision making efficiency decision node; and a workload model responsive to an increase team decision node and to the system performance and operator performance models for providing a decision signal representing whether to change the size of the team of unmanned vehicles.

7 . The decision tool of claim 6 including:
a utility node for each decision node;
multiple nature nodes interactive with each other and with the utility nodes including observable nature nodes representative of one or more node indications from the node group of Total Task Time, Wait Times due to Loss of Situation Awareness (WTSA), Frequency of Reassignment, ID Task Success Rate, Neglect Time (NT), System Interruption, UV Health Status, Utilization Time (UT), Interaction Times (IT), Wait Times due to Queue (WTQ), Elimination Task Success Rate, Total time to Target Elimination, Team Heterogeneity and Team Size nodes;
unobservable nature nodes representative of one or more node indications from the group of System Performance, Operator Performance, Workload, Situation Awareness (SA), Information Overload, Automation level, Enemy Team, Task Complexity and Task Heterogeneity nodes; and
a nature node representative of Task Complexity which serves as a redundant node.

8 . The decision tool of claim 7 wherein the decision tool is a Bayesian Belief network.

9 . In a system for controlling a team of unmanned vehicles, a computer model network decision tool comprising:
a sequence of decision nodes in a hierarchical order representing a hierarchical decision making process in which later decisions depend on earlier decisions; a series of observable parent nature nodes that provide optimal input signals to optimize each decision node through the network, the observable parent nature nodes related to the decision nodes; a series of unobservable nodes that are computed from the observable nodes and are part of the output that the model provides to represent whether to change the size of the team of unmanned vehicles.

10 . The decision tool of claim 9 comprising:
a utility node for each decision node;

the series of nature nodes interactive with each other and with the utility nodes, the nature nodes including one or more decision node indications from the group of Interface Usability; Adequate Automation; Algorithm Efficient; Tasks Management Efficiency; Operator Decision Making (DM) Efficiency; and Increase Team decision nodes.

11 . The decision tool of claim 10 wherein the utility nodes include one or more utility node indications from the group of Utility Interface; Utility Automation; Utility System DM; Utility Task Management; Utility Operator DM; and Utility of Capacity utility nodes.

12 . The decision tool of claim 11 wherein the observable nature nodes include one or more observable nature node indications from the group of Wait Times due to Loss of Situation Awareness (WTSA); Neglect Time (NT); Interaction Time (IT); Wait Times due to Queue (WTQ); Utilization Time (UT); System Interruption; Elimination Task Success; UV Health Status; Total Time for Target Elimination; Team Size; Team Heterogeneity; Frequency of Reassignment; Total Task Time; and ID Success Rate nodes.

13 . The decision tool of claim 12 wherein the unobservable nodes include one or more unobservable nature node indications from the group of System Performance; Information Overload; Situation Awareness (SA); Workload; Wait Times due to Cognitive Reorientation; Automation Level; Operator Performance; Task Complexity; Enemy Team; and Task Heterogeneity unobservable nature nodes.

Assignees

• Rodas Maria Olinda
• Us Navy

Inventors

• Rodas Maria Olinda

Classifications

• A61B5/18

  for vehicle drivers {or machine operators} · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title

• G05D1/0027 Primary

  involving a plurality of vehicles, e.g. fleet or convoy travelling (fleet control of land vehicles from a control room G05D1/0297; traffic control systems for road vehicles G08G1/00; for marine craft G08G3/00; for aircraft G08G5/00) · CPC title