Cognitive tour guide system

Patent metadata

Field	Value
Publication number	US-2017211940-A1
Application number	US-201615005112-A
Country	US
Kind code	A1
Filing date	Jan 25, 2016
Priority date	Jan 25, 2016
Publication date	Jul 27, 2017
Grant date	—

Abstract

Methods, computer program products, and systems are presented. The methods include, for instance: providing a cognitive tour guide service to a group of participants for a tour with an initial route planned by participants registration information and environment information along the initial route. During the tour, real time sensory data on the participants health characteristics, change in environment are collected by a cognitive agent accompanying the group to lead the tour are relayed to a cognitive tour guide engine, and real time multi-objective optimization is modeled and performed. The participants are regrouped responsive to their respective enjoyment and circumstances of the environment and, among a set of optimal solutions, a new route for the tour is selected based on participant preference modeled and trained by participant feedback. During the tour, the cognitive tour guide engine continuously optimizes route responsive to incoming real time sensory data.

First claim

1 . A computer implemented method for providing a cognitive tour guide service to a group of participants, comprising:
collecting, by a cognitive tour guide engine running on a computer, sensory data on the participants and on environments of a tour in real time captured by two or more cognitive agents accompanying and leading the group for the tour with a present route such that the cognitive tour guide engine learns location of the group, respective pace of travel for the participants, and respective level of interest of the participants for a landmark in the present route; optimizing, by the cognitive tour guide engine, the present route by use of at least two objectives of the tour in real time during the tour, a first objective of the tour from the at least two objectives being to minimize expenses for the tour, and a second objective of the tour from the at least two objectives being to maximize enjoyment of the participants, based on the collected real time sensory data; and notifying, by the cognitive tour guide engine, the participants and the two or more cognitive agents of two or more new routes generated from the optimizing in real time, respective new routes of the two or more new routes corresponding to each subgroup of two or more subgroups formed from the group of participants based on each objective of the at least two objectives of the tour, such that the participants of each subgroup may continue the tour according to the respective new route corresponding to respective subgroups.

2 . The method of claim 1 , further comprising:
registering a first participant among the participants for the tour with an initial route prior to the tour, the initial route being calculated from individual participant registration data as provided by the first participant, the participant registration data comprising face image, body measurements, and data for health characteristics.

3 . The method of claim 1 ,
wherein the two or more cognitive agents are respective unmanned aerial vehicles capable of real time sensory data collection and real time data communication, wherein the sensory data comprises visual data, auditory data, and/or gestural data of the participants captured by respective devices installed on each cognitive agent, wherein each cognitive agent of the two or more cognitive agents communicates in real time with the cognitive tour guide engine, other cognitive agents, and a mobile communication terminal of respective participants in the group, and wherein said each cognitive agent relays data stream from the mobile communication terminal to the cognitive tour guide engine, the data stream being generated by an applet installed on the mobile communication terminal.

4 . The method of claim 1 ,
wherein the sensory data on the environment of the tour comprises topographical scan of the present route, obstacles on the present route, weather, delays of the tour caused by traffic, concentration of visitors in a landmark in the present route and/or an emergency to the participants.

5 . The method of claim 1 , said collecting comprising:
obtaining the sensory data on the participants and on the environments of the tour from a first cognitive agent of the two or more cognitive agents; and identifying a level of interest of a participant among the participants on the landmark by measuring time spent by the participant at the landmark.

6 . The method of claim 1 , said optimizing comprising:
determining a Pareto optimal frontier for the at least two objectives of the tour, the Pareto optimal frontier specifying combinations of trade-offs between the at least two objectives; generating a set of Pareto optimal solutions satisfying the at least two objectives as much as possible, by executing the determined Pareto optimal frontier; building a stochastic model of participant preference for the generated set of Pareto optimal solutions; selecting a dominant solution amongst the set of Pareto optimal solutions based on the stochastic model of participant preference; gathering feedback from the participants on the selected dominant solution upon informing the participants of the selected dominant solution to train the stochastic model of participant preference in subsequent cycles of the optimizing; and regrouping the participants into the two or more subgroups and creating the two or more new routes respectively for the two or more subgroups.

7 . The method of claim 1 , further comprising:
leading a first subgroup of the two or more subgroups, by a first cognitive agent of the two or more cognitive agents, based on a first new route of the two or more new routes, responsive to the notifying.

8 . A computer program product comprising:
a non-transitory computer storage medium readable by one or more processor and storing instructions for execution by the one or more processor for performing a method for providing a cognitive tour guide service to a group of participants, comprising:
collecting, by a cognitive tour guide engine running on a computer, sensory data on the participants and on environments of a tour in real time captured by two or more cognitive agents accompanying and leading the group for the tour with a present route such that the cognitive tour guide engine learns location of the group, respective pace of travel for the participants, and respective level of interest of the participants for a landmark in the present route;
optimizing, by the cognitive tour guide engine, the present route by use of at least two objectives of the tour in real time during the tour, a first objective of the tour from the at least two objectives being to minimize expenses for the tour, and a second objective of the tour from the at least two objectives being to maximize enjoyment of the participants, based on the collected real time sensory data; and
notifying, by the cognitive tour guide engine, the participants and the two or more cognitive agents of two or more new routes generated from the optimizing in real time, respective new routes of the two or more new routes corresponding to each subgroup of two or more subgroups formed from the group of participants based on each objective of the at least two objectives of the tour, such that the participants of each subgroup may continue the tour according to the respective new route corresponding to respective subgroups.

9 . The computer program product of claim 8 , further comprising:
registering a first participant among the participants for the tour with an initial route prior to the tour, the initial route being calculated from individual participant registration data as provided by the first participant, the participant registration data comprising face image, body measurements, and data for health characteristics.

10 . The computer program product of claim 8 ,
wherein the two or more cognitive agents are respective unmanned aerial vehicles capable of real time sensory data collection and real time data communication, wherein the sensory data comprises visual data, auditory data, and/or gestural data of the participants captured by respective devices installed on each cognitive agent, wherein each cognitive agent of the two or more cognitive agents communicates in real time with the cognitive tour guide engine, other cognitive agents, and a mobile communication terminal of respective participants in the group, and wherein said each cognitive agent relays data stream from the mobile communication terminal to the cognitive tour guide engine, the data stream being generated by an applet installed on the mobile communication terminal.

11 . The computer program product of claim 8 ,
wherein the sensory data on the environment

Assignees

• IBM

Inventors

• Baughman Aaron Keith
• Eggenberger Christian
• Jenzer Lars
• Trim Craig Matthew

Classifications

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title

• A61B5/01

  Measuring temperature of body parts {; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue} (clinical contact thermometers G01K13/20) · CPC title

• A61B3/113

  for determining or recording eye movement · CPC title

• G06N20/00

  Machine learning · CPC title