Method and system for calibration of a space in an indoor area

1. A server computing device for generating a fingerprint map for an indoor area, where layout of the indoor area is previously unknown, the server computing device comprising:

a processor; and

a memory storing a set of instructions, the instructions executable in the processor to:

obtain, from one or more data recording devices, sensorial data gathered for a space to be mapped in the indoor area;

derive a plurality of trajectories from the sensorial data, the plurality of trajectories indicating movement of the one or more data recording devices in the space;

estimate a plurality of map constraints of the space by applying geometry-based zone detection technique on the plurality of trajectories, the plurality of map constraints including at least a non-accessible zone;

create, based on the plurality of trajectories and the plurality of map constraints, a structural map representing the space;

extract, from the sensorial data, calibration information associated with a series of locations in the space; and

generate, based on the calibration information and the structural map, a fingerprint map of the space for localization inside the space.

2. The server computing device of claim 1 further comprising instructions to trigger each of the one or more data recording devices to automatically commence gathering of the sensorial data in the space, based on a location of each of the one or more data recording devices in relation to the space.

3. The server computing device of claim 2 , wherein the sensorial data comprises at least one of data from inertial sensors and data from signal sensors of each of the one or more data recording devices.

4. The server computing device of claim 1 , wherein the plurality of map constraints of the space comprise at least one of accessible zones in the space, a doorway of the space, and points of transition of elevation inside the space.

5. The server computing device of claim 4 , wherein the accessible zones comprise a pedestrian area and one of a pathway and an open passageway area, and the non-accessible zones comprise one of a wall and a floor opening.

6. The server computing device of claim 1 further comprising instructions executable in the processor to:

estimate, based on the plurality of trajectories, a plurality of anchor points inside the space, each of the plurality of anchor points indicating one of a site of transition in an elevation inside the space and a site of transition between the indoor area and an outdoor area; and

select a set of relevant trajectories from amongst the plurality of trajectories connecting the plurality of anchor points inside the space, the set of relevant trajectories being selected based on a confidence score of each trajectory connecting the plurality of anchor points.

7. The server computing device of claim 6 , wherein the confidence score of each trajectory is determined based on at least one of: a theoretical best fit between the anchor points, a similarity of the trajectory with a majority proportion amongst the plurality of trajectories, and an auxiliary corroborating technique.

8. The server computing device of claim 7 , wherein the auxiliary corroborating technique comprises employing signal from at least one of a satellite-based navigation system and a cellular signal-based locating system.

9. The server computing device of claim 1 further comprising instructions executable in the processor to:

estimate, based on the plurality of trajectories, a plurality of concentrated anchor points in vicinity of at least one location inside the space; and

fuse the plurality of concentrated anchor points to estimate an anchor point at the at least one location inside the space, wherein the anchor point is indicative of one of a site of transition in an elevation inside the space and a site of transition between the indoor area and an outdoor area.

10. The server computing device of claim 9 further comprising instructions executable in the processor to fuse the plurality of concentrated anchor points to estimate the anchor point using a machine-learning based clustering technique.

11. A method for generating a fingerprint map for an indoor area, where layout of the indoor area is previously unknown, the method comprising:

obtaining, from one or more data recording devices, sensorial data gathered for a space to be mapped in an indoor area;

deriving a plurality of trajectories from the sensorial data, the plurality of trajectories indicating movement of the one or more data recording devices in the space;

estimating a plurality of map constraints of the space by applying geometry-based zone detection technique on the plurality of trajectories, the plurality of map constraints including at least a non-accessible zone;

creating, based on the plurality of trajectories and the plurality of map constraints, a structural map representing the space;

extracting, from the sensorial data, calibration information associated with a series of locations in the space; and

generating, based on the calibration information and the structural map, a fingerprint map of the space for localization inside the space.

12. The method of claim 11 , wherein the obtaining the plurality of trajectories comprises triggering each of the one or more data recording devices to automatically commence gathering of the sensorial data in the space, based on a location of each of the one or more data recording devices in relation to the space, the sensorial data comprising at least one of data from inertial sensors and data from signal sensors of each of the one or more data recording devices.

13. The method of claim 12 , wherein the triggering comprises determine a transition of the one or more data recording device from an outdoor area to the indoor area using at least one of a satellite-based navigation system and a cellular signal-based locating system.

14. The method of claim 11 , wherein the plurality of map constraints of the space comprise at least one of accessible zones in the space, a doorway of the space, and points of transition of elevation inside the space.

15. The method of claim 14 , wherein the accessible zones comprises a pedestrian area and one of a pathway and an open passageway area, and the non-accessible zones comprises one of a wall and a floor opening.

16. The method of claim 11 , wherein the deriving comprises:

estimating, based on the plurality of trajectories, a plurality of anchor points inside the space, each of the plurality of indicating a site of one of a transition in an elevation inside the space and a transition between the indoor area and an outdoor area; and

refining trajectories by selecting a set of relevant trajectories from amongst the plurality of trajectories connecting the plurality of anchor points inside the space, the set of relevant trajectories being selected based on a confidence score of each trajectory connecting the plurality of anchor points.

17. The method of claim 16 , wherein the confidence score of each trajectory is determined based on at least one of: a theoretical best fit between the anchor points, a similarity of the trajectory with a majority proportion amongst the plurality of trajectories, and an auxiliary corroborating technique.

18. The method of claim 17 , wherein the auxiliary corroborating technique comprises employing signal from at least one of a satellite-based navigation system and a cellular signal-based locating system.

19. The method of claim 11 wherein the deriving comprises:

estimating, based on the plurality of trajectories, a plurality of concentrated anchor points in vicinity of at least one location inside the space; and

fusing the plurality of concentrated anchor points to estimate an anchor point at the at least one location inside the space, wherein the anchor point is indicative of one of a site of transition in an elevation inside the space and a site of transition between the indoor area and an outdoor area.

20. The method of claim 19 , wherein the fusing the plurality of concentrated anchor points to estimate the anchor point is performed using a machine-learning based clustering technique.