Claims:We Claim:
We claim:
1. A laptop position detection system (500) for determining the position of usage of a laptop (100), said laptop position detection system (500) comprising at least:
- a sensor (200) connected to the base of said laptop (100);
characterized in that
at least one control unit (300) that takes input from said sensor (200) for
- calculating an angle of base (105) of said laptop (100) using continuous signals (205) from said sensor (200),
- determining at least one status of the position of usage of said laptop (100) as “In bag” based on said angle of base(105),
- extracting features (330) of said continuous signals (205) obtained from said sensor (200) by filtering said continuous signals (205) using filters (305,310,315) and passing said filtered signal (325) through a time synchronizer block(320),
- determining at least one status of the position of usage of said laptop (100) as “On lap” or “On table” based on said extracted features (330) of said filtered signal (325) using a decision tree based machine learning model (340),
- providing the final output for said laptop(100) based on said status of the position of usage of said laptop (100) using a state machine (350) and at least one timer (355).

2. The laptop position detection system (500) as claimed in claim 1, wherein said sensor (200) is a 3 axis accelerometer sensor.

3. A method for determining the position of usage of a laptop (100) using a laptop position detection system (500), said laptop detection position system (500) comprising a sensor (200) with a control unit (300), said method comprising the steps of:

- calculating an angle of base (105) of said laptop (100) using continuous signals (205) from said sensor (200),
- determining at least one status of the position of usage of said laptop (100) as “In bag” based on said angle of base(105),
- extracting features (330) of said continuous signals (205) obtained from said sensor (200) by filtering said continuous signals (205) using filters and passing said filtered signal through a time synchronizer block(320),
- determining at least one status of the position of usage of said laptop (100) as “On lap” or “On table” based on said extracted features (330) of said filtered signals (325) using a decision tree based machine learning model(340),
- providing the final output (360) for said laptop(100) based on said status of the position of usage of said laptop (100) using a state machine (350) and at least one timer (355).

4. The method as claimed in claim 3, wherein said angle of said laptop base (105) is derived from pitch value (110) and roll value (115) obtained from said sensor (200), said pitch value (110) and roll value (115) triggered by continuous signals (205) from said sensor (200).

5. The method as claimed in claim 3, wherein said “In bag” status position of usage of said laptop (100) is based on said calculated angle of said laptop base (105) and motion of said laptop (100).

6. The method as claimed in claim 5, wherein said calculated angle of said laptop (100) is detected using a combination of said pitch value (110) and roll value (115) computed from continuous signal (205) obtained from said sensor (200) and determining said motion of said laptop (100) as stationary.

7. The method as claimed in claim 3, wherein said extraction of features of said continuous signals (205) is done by passing said continuous signals through high pass filters (305) and band pass filters (310) along with filtering said pitch value (110) and roll value (115) computed from said continuous signals (205) from said sensor (200) using filter banks (315) and passing collective filtered signal (325).

8. The method as claimed in claim 7, wherein said pitch value (110) and roll value (115) are filtered through said filter banks (315) to decompose them into different frequency bands (335) and passing said frequency bands through a time synchronizer block (320).

9. The method as claimed in claim 8, wherein said frequency bands (335) obtained by passing said pitch value (110) and roll value (115) through said filter banks(315) and said signals (205) obtained from said sensor (200) are time aligned using said time synchronizer block (320) since the group delays of different filters may not be same.

10. The method as claimed in claim 3, wherein said decision tree based machine learning model (340) is trained offline with pre-collected samples of sensor (200) data for various ways of using said laptop (100) in “On lap” and “On table” position.

11. The method as claimed in claim 3,wherein said “In bag” or “On lap” or “On Table” status of usage position of said laptop (100) obtained as output (345) from said decision tree based machine learning model (340) is further fed as input in said state machine (350).

12. The method as claimed in claim 10, wherein said state machine(350) after receiving consecutive said status of usage position of said laptop (100) for a certain period of time measured by said timer(355) produces said final output (360).

, Description:Complete Specification:

The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed.
Field of the invention
[0001] The present disclosure relates to a laptop position detection system.
Background of the invention
[0002] Using laptops on the lap for an extended period of time is usually not ideal and neither comfortable owing to the excessive heat that is generated by the battery. It is also preferable for users that the laptop be connected to wireless internet connections rather than wired Ethernet which can restrict the freedom of users from working from any place or restrict their movement as such. The heat generated as well as the radiations emitted from the laptop can also cause long-term health issues.
[0003] The power drainage of laptops is also unregulated and it is constant no matter how the laptop is used or whether it is in use at all. Thus no matter if the laptop is in sleep mode in the bag or left idle on the table or in use by the user, the power drainage remains constant thereby creating a lot of inconvenience for the use. This can not only lead to reducing the life of the laptop battery because of having to put on charge constantly but can also diminish user experience.

Brief description of the accompanying drawings
[0004] An embodiment of the disclosure is described with reference to the following accompanying drawings.
[0005] Figure 1 illustrates a laptop highlighting the base of said laptop (100) where the laptop position detection system (500) would reside along with showcasing a pitch value (110) and a roll value (115) of said laptop (100) which are used for the purposes of the embodiment.
[0006] Figure 2 illustrates a view of the base (105) of said laptop (100) that contains the tangible parts of said laptop position detection system (500).
[0007] Figure 3 illustrates a block diagram showcasing the detection of “On Lap” or “On Table” position of laptop (100) using our invention.
[0008] Figure 4 illustrates a block diagram showcasing the further detection of “In Bag” position of laptop (100) using our invention.
[0009] Figure 5 illustrates a block diagram showcasing the final output (360) of our invention.

Detailed description of the drawings
[0010] Figure 1 illustrates a laptop (100).The base of (105) said laptop makes space for the tangible parts of a laptop position detection system (500). The laptop position detection system (500) makes use of inputs in the form of pitch value (110) and roll value (115) which is further used to calculate the angle of said laptop base (105) from the surface on which the laptop rests.
[0011] Figure 2 illustrates the tangible portions of the laptop position detection system (500) as is affixed on the base (105) of said laptop (100). The tangible portions include at least one accelerometer sensor (300) and at least one control unit (300).
[0012] The accelerometer sensor (200) in said laptop (100) provides continuous signals (205) to the control unit (300) that further determines the pitch value (110) and roll value (115) of said laptop base (105) from the surface on which the laptop rests. The calculated pitch value (110) and roll value (115) is then further used to determine the angle of laptop base (105) of said laptop (100).
[0013] The control unit (300) of the laptop position detection system does the computations using algorithms and machine learning model (340) to give a final output (360) based on which the usage of the laptop (100) can be optimized.
[0014] One initial detection of laptop (100) position is based on said calculated angle of base (105) of said laptop (100) from the continuous signals (205) of said accelerometer sensor (200) along with a combination of motion of said laptop (100). This combination is used to determine the “In Bag” status of said laptop (100).
[0015] The “In Bag” status detection can further classify the status as whether the laptop (100) is inside a backpack or inside a sling bag. In case of laptop being inside a backpack, the pitch angle (110) is not considered and in case of laptop () being in a sling bag, the roll angle (115) is not considered.
[0016] In one embodiment of the invention, said accelerometer sensor (200) is a 3-axis MEMS accelerometer sensor.
[0017] Figure 4 illustrates a block diagram to extract the features (330) of said continuous signals (205) obtained from said accelerometer sensor (200) for use in further status computations.
[0018] The continuous signals are used to estimate pitch value (110) and roll value (115). The pitch value (110) and roll value (115) are further decomposed into different frequency bands of interest (355) using filter banks (315). This decomposition is necessary for signal analysis engine to handle wide variety of scenarios for status determination of laptop (100).
[0019] The continuous signals (205) are also filtered with high-pass (305) and band-pass filters (310) for more detailed analysis. The design of high-pass (305) and band-pass filters (310) for continuous signal (205) and the filter banks (315) for pitch value (110) and roll value (115) are based on empirical computation on signal database, which are collected after fixing 3-axis accelerometer sensors (200) on laptop’s base (105) panel. Since the group delays of different filters may not be same, the filtered signals (325) are time-aligned in Time Synchronizer block (320).
[0020] In one of our embodiment, the cut-off frequency of high pass filter (305) is set as 30Hz. Similarly for band-pass filter (310), the pass-band frequencies are set from 1Hz to 20Hz.
[0021] Figure 4 illustrates a block diagram showcasing the determination of status of laptop (100) as “On table” or “On Lap” based on using said extracted features (330) of continuous signals (205) and using that as input for a decision tree machine learning model (340) that produces said output (345). This model (340) is trained offline with pre-collected samples of accelerometer data for various ways of using laptop in on lap and on table position.
[0022] Figure 5 illustrates a state machine, said state machine (350) after receiving consecutive said status of usage position of said laptop (100), whether “In Bag” or “On Table” or “On lap”, for a certain period of time measured by said timer (355) produces said final output (360).
[0023] The timer (355) is used to establish the transition between each state. If In-Bag detection engine detects position as ‘In-Bag’, then state machine module will be disabled and provide the output as ‘in-Bag’. However if the output is ‘out-of-Bag’, then state machine module will get enabled. This module’s function is to introduce a stability in the outputs determined by machine learning model and to avoid the frequent toggling between the outputs.
[0024] It must be understood that the embodiments explained in the above detailed description are only illustrative and do not limit the scope of this invention. Any modification to the laptop position detection system are envisaged and form a part of this invention. The scope of this invention is limited only by the claims.