`BEE”
`INTELLECTUAL
`PROPERTY DNDDA
`diafisnsm mm
`mm min
`
`GOVERNMENT OF INDIA
`
`am na‘ 333th m
`MINISTRY OF COMMERCE & INDUSTRY
`
`flit W
`THE PATENT OFFICE
`
`TO WHOMSOEVER IT MAY CONCERN
`
`WWWWafimfimmaafim$fimmmg
`magmfifimfivffifimfififiafieemérmfimamwm
`afiefigfififlmmmfi
`
`(mi? WWAuthorised ignatory)
`
`#Wardeemm 7970567a7w73(3) 3%me
`
`l, the undersigned, being an officer duly authorized to sign and issue the certificate on
`
`behalf of the Controller General of Patents, Designs and Trademarks in accordance with the
`
`provisions of Section 73(3) of the Patents Act, 1970, hereby certifi/ that annexed hereto is
`
`aTrue Copy ofthe document(s] asfiled in connection with thefollowing PatentApplication:
`
`as) 317733? man/a) Application Number: 201741039669
`
`z?) 17573? an?) #3 37—02%) Date ofFiIing: 07/11/2017
`
`Jr) W Wm?) 657 m-
`C) Name of the document(s) requested: Provisional Specification
`
`deem 7970 «#9777 747(1) $£€flfigsifififéfielfi7wi$agfimfl
`is certificate is issued under the powers vested in me U/S 147(1) of The Patents
`
`'
`
`"5 81h day of November 2018
`
`@2733? ll??? WControlleroj‘ngnts
`
`
`
`FORM 1
`(FOR OFFICE USE ONLY)
`
`
`THE PATENTS ACT, 1970 (39 of 1970)
`&
`
`
`
`The Patents Rules, 2003
`APPLICATION FOR GRANT OF
`PATENT
`
`
`
`[See section 7,54 & 135 and sub- rule (1) ofrule
`20]
`
`
`
`CBR No:
`
`l. APPLICANT’S REFERENCE/
`IDENTIFICATION NO. (AS
`ALLOTTED BY OFFICE)
`
`
`
`
`
`
`
`Patent of
`
`Addition( )
`
`
`
`3A. APPLICANT(S)
`Name in Full
`Address of the
`A . licant
`UBER
`1455 Market Street,
`
`
`San Francisco,
`TECHNOLOGIES,
`
`
`
`
`
`lNC.
`California 94103 ,
`
`
`United States of
`
`
`.
`America
`
`
`
`33. CATEGORY OF APPLICANT [Please tick (V ) at the appropriate category]
`
`2. TYPE OF APPLICATION [Please tick ( \/ ) at the appropriate category]
`
`Ordinary ([2!)
`
`Convention (El)
`
`PCT-NP (a )
`
`Divisional
`( )
`_
`
`Patent of
`Addition( )
`
`Divisional
`(
`)‘
`
`Patent of
`Addition( )
`
`Divisional
`( )
`
`Nationality
`
`'
`
`United States
`
`Country of
`Residence
`United
`States of
`America
`
`
`
`Natural Person (l2!)
`
`Other than Natural Person
`
`PD027275iN-SC
`
`
`
`
`
`Small Entity (El)
`
`4. INVENTOR(S Please tick (/ ) at the appropriate category]
`
`‘ re all the inventor(s) same as Yes ( )
`No (El)
`he applicant(s) named above?
`V
`f “No”, furnish the details of the inventor(s)
`
`
`
`
`
`
`
`Name in Full
`
`Nationality
`
`CHOPRA, Naomi
`
`Indian
`
`Country of
`Residence
`
`Address of the Inventor
`
`
`
`
`
`
`
`
`1455 Market Street, San
`United
`States of
`Francisco, California
`
`
`
`America
`94103, United States of
`
`
`America
`
`
`
`
`
`United
`1455 Market Street, San
`States of i
`, Francisco, California
`
`America
`I 94103, United States of
`
`' America
`
`
`6. AUTHORISED
`
`
`REGISTERED PATENT
`
`
`PRASHANT PHILLIPS
` Mobile No.
`98 183 01173
`
`LAKSHMI KUMARAN &
`7. ADDRESS FOR
`
`SERVICE OF APPLICANT
`IN INDIA
`SRI DHARAN _
`
`
`
`Postal Address
`
`(+91) 044 2833 4700
`
`ANNAMALAI,
`Sundar
`
`Indian
`
`5. TITLE OF THE INVENTION
`
` MAPS ANOMALY DETECTION SYSTEM
`
`IN/PA-1229
`
`AGENT(S)
`
`2, Wallace garden, 2nd Street,
`Chennai — 600 006
`India
`
`
`
`
`98183 01173
`I
`
`
`
`
`
`
`
`
`8. IN CASE OF APPLICATION CLAIMING PRIORITY OF APPLICATION
`
`
`FILED IN CONVENTION COUNTRY, PARTICULARS OF CONVENTION
`APPLICATION : N/A
`
`PD027275IN-SC
`
`
`
`Country
`
` Application
`
`Number
`
`
`Filing Date Name of
`Applicant
`
`Nil
`
`Nil
`
`Nil
`
`Nil
`
`Nil
`
`Title of
`Invention
`
`
`
`[PC (as
`classified in the
`convention
`
`country)
`
`
`
`
`
`
`
`9.
`
`IN CASE OF PCT NATIONAL PHASE APPLICATION, PARTICULARS OF
`
`INTERNATIONAL APPLICATION FILED UNDER PATENT CO-OPERATION
`
`TREATY (PCT) : N/A
`_
`
`.
`
`,
`
`International filing date as allotted by the
`
`
`
`10.
`
`IN CASE OF DIVISIONAL APPLICATION FILED UNDER SECTION
`
`16, PARTICULARS OF ORIGINAL (FIRST) APPLICATION : N/A
`
`Original (First) Application No.
`
`Date of filing of Original (First)
`AM.--
`
`Nil
`
`'
`
`Nil
`
`I]. IN CASE OF PATENT OF ADDITION FILED UNDER SECTION 54,
`
`
`PARTICULARS OF MAIN APPLICATION OR PATENT: N/A
`
`Date of filing of Main Application
`
`Main Application/Patent N0.
`
`‘
`
`
`
`Nil
`
`Nil
`
`12. DECLARATIONS
`
`(i) Declaration by the Inventor(s)
`(In case the applicant is an assignee: the inventor(s) may sign herein below or the
`applicant may upload the assignment or enclose the assignment with this
`application for patent or send the assignment by post/electronic transmission duly
`authenticated within the prescribed period).
`I/We, the above named inventor(s) is/are the true & first inventor(s) for this invention
`and declare that the applicant(s) herein is/are my/our assignee or legal representative.
`
`ANNAMALAI, Sundar
`
`PDO27275lN-SC -
`
`
`
`(ii) Declaration by the Applicant(s) in the convention country
`
`(In case the applicant in India is different than the applicant in the convention
`country: the applicant
`in'the convention country may sign herein below or
`applicant
`in India may upload the assignment
`from the applicant
`in the
`convention country or enclose the said assignment with this application for
`patent or send the assignment by post/electronic transmission duly authenticated
`within the prescribed period)
`
`I/We, the Applicant(s) in the convention country declare that the applicant(s) herein
`is/are my/our assignee or legal representative.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(iii) Declaration by the Applicant(s)
`[/We the applicant(s) hereby declare that
`[Z I am /We are in possession of the above mentioned invention.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`[2] Specification relating to the invention is filed with this application.
`[2| The invention as disclosed in the specification uses the biological material from
`India and the necessary permission from the competent authority shall be
`submitted by me/us before the grant of patent to me/us.
`[2! There is no lawful ground of objection(s) to the grant of patent to me/us.
`E!
`I am / we are the true & first Inventors.
`
`1 am / we are the assignee or legal representative of true & first Inventors.
`ll
`The application or each of applications, particulars of which are given in
`[E
`
`Paragraph-8 was the first application in convention country / countries in respect
`
`PD0272751N-SC
`
`
`
`I3
`
`of my/our invention(s).
`I / We claim priority from the above mentioned application(s) filed in convention
`country/countries and state that no application for protection in respect of the
`invention has been made in a convention country before that date by me/us or by
`
`in or modification of the invention
`The said invention is an improvement
`particulars of which are given in Paragraph-1 l.
`i
`13. FOLLOWING ARE THE ATTACHMENTS WITH THE APPLICATION
`
`any person from which l/We derive the title.
`My/our application in India is based on International application under Patent
`Cooperation Treaty (PCT) as mentioned in Paragraph-9.
`The application is divided out of my/our application particulars of which is given
`in Paragraph-10 and pray that this application may be treated as deemed to have
`been filed on N/A under section 16 of the Act.
`
`
`
`(a) Form 2
`
`I Complete/
`I Provisional
`
`No. of pages of
`Description :27
`
`specification
`(Description Part)#
`
`Claim(s)
`
`.ofcl-aims : 0
`
`. of pages : 0
`
`Abstract
`
`. of pages : O
`
`Drawing(s)
`
`. of drawings : 5
`
`. of pages : 5
`
`the
`if the applicant desires to adopt
`# In case of a complete specification,
`drawings filed with his provisional specification as the drawings or part of the
`drawings for the complete specification under
`rule 13(4),
`the number of
`such pages
`filed with the provisional
`specification are required to be
`mentioned here.
`
`PD027275IN-SC
`
`
`
`(b) Provisional Specificationékpeeafematien-Mth—theéntemaéenal-appfieafienlafi
`
`
`
` (g) Statement and undertaking on FORM-3
`
`(h) Power of Authority—To follow
`
`
`
`Total fee *9,600/- in Cash/ Banker's Cheque [Bank Draft bearing No
`........................................... Date
`on HDFC BANK LTD.
`
`
`
`
`
`
`
`
`
`
`
`
`l/We hereby declare that to the best of my/our knowledge, information and belief the fact
`and matters stated herein are correct and l/We request that a patent may be granted to
`
`_ me/us for the said invention.
`
`Date: 07 November 2017
`
`Signature:
`
`Name: PRASHANT PHILLIPS
`lN/PA—1229
`0F LAKSHMI KUMARAN & SRIDHARAN
`
`AGENT FOR THE APPLICANT(S)
`
`To,
`The Controller of Patents
`The Patent Office at Chennai
`
`
`PD0272751N-SC
`
`
`
`FORM 2
`
`THE PATENTS ACT, 1970
`
`(39 of 1970)
`&
`
`THE PATENTS RULES, 2003
`
`PROVISIONAL SPECIFICATION
`See section 10, rule 13
`
`I. Title ofthe invention: MAPS ANOMALY DETECTION SYSTEM
`
`070\"(m03%“(6(9<70?\l10W“.
`
`ADDRESS
`1455 Market Street, San Francisco,
`California 94103, United States of
`America
`
`2. Applicant(s)
`NAME
`UBER TECHNOLOGIES,
`INC.
`
`NATIONALITY
`United States
`
`3. Preamble to the description
`
`PROVISIONAL SPECIFICATION
`
`The following specification describes the invention.
`
`PD02727SIN-SC
`
`
`
`Maps Anomaly Detection system:
`Concept
`"
`'
`'
`‘
`
`Maps Anomaly Detection (MAD) system employs a set of algorithms to programmatically detect
`v anomalies in maps data (including missing roads, incorrect road connections,- incorrect
`geometry, incorrect and'missing turn restrictions) with high accuracy, withoUt manual
`intervention u'sing actual GPS traces and predicted routes from A to B._
`-
`
`Concept in brief
`
`.
`MAD'takes the following information as input
`1. Raw GPS trip trace for each trip ( actual route of the trip as taken by the driver). — Series
`of lat/longs
`,
`(
`_
`’
`2. Suggested roUte of the map provider you want to find issues in (routing algorithm is
`assumed to be optimal) - Series‘of lat/longs.
`
`.
`
`Using these two sets of information for a number of trips, we compare the actual route taken by
`the driver with the route predicted by a_routing engine based on the provider's maps data and
`compute the difference between the two routes. .-
`
`
`
`
`
`This issue has occurred more than 500 times
`on the day we ran the tool (0.5% of the trips).
`
`l
`
`ACTUAL RGJTE TM£N
`EV fi-E DRIVER
`
`
`
`
`MAD employs the following process, each step is its own algorithm as discussed in the following
`section.
`
`INPUT DATA
`
`INTERPOLATE
`Make lat/longs equidistant in the route.
`
`not in me other
`
`Compute palms present in one route but
`
`FILTER DRIVER BEHAVIOUR/6P5
`NOISE
`
`‘5
`*
`AGGREGATE RESULTS
`Combine multiple anomalous paints to l
`
`output issues
`3
`
`CLUSTER GPS POINTS lNTO lSSU S
`All 995 points of one issue are combmad
`inm one cluster
`
`
`
`
`
`MAP ISSUES
`
`Algorithm
`
`interpolation
`
`Both the polylines are interpolated so that the distance between any pair of adjacent
`points is less than a threshold (5 meters in our analysis).
`
`
`
`BEFORE iNTERPOLATION
`
`AFTER INTERPOLATION
`
`Route Diff computation
`
`We then find the lat/longs which are present in one of the routes but not present in the other
`)
`route.
`in order to find it a lat/long is present in the other route, we Check it there is any other point in
`the other route within a threshold radius (15 meters in our analysis) to the existing point. We doi
`,3.
`this for each lat long in both the routes and just keep those that are not present in the diff.
`
`/ .
`
`
`
`
`“ring-3116'“; ta; ‘ "‘ 813st:
`
`
`
`
`
`
`43’ .‘o. ,‘G—jfl
`
`
`
`
`
`nahisodtg
`. 1» n’wn.
`'
`
`Mam
`
`‘ “ “ \'Z'it-3\Mi§t‘"‘mao ‘
`
`
`
`
`
`«A tum; an ,
`
`swwanpaiyu
`
`,.
`
`'
`
`'
`
`cum.
`,
`mmimmum ._
`‘ 15:0”
`- mver
`‘
`
`V '
`
`‘ V
`1
`
`a“
`
`an '12?m
`
`”guy:
`3,...“
`
`;:
`3
`2
`..
`2
`«vi-ma 10‘
`, rum.
`hm
`s
`1-. .
`
`/
`
`,
`
`»
`
`__
`
`.
`
`«
`
`.
`
`3"
`t. - uw fi'
`'
`WV
`.. 3
`42:31 §
`any:
`I
`.
`" V
`
`
`«
`
`VACTUAL MENUS PREDICTED and PREDICTFD MINUS ACTUAL
`
`U1
`
`
`
`Filtering out driver behavioUr/GPS noise
`
`\
`
`One we have the 2 set of diffs, we then filter out those lat/longs which occurred in the diff due to
`reasons other than map data issue. This could be driver behaviour or gps noise.
`
`‘
`
`1.. Pair the sections in “actual minus predicted" with the corresponding sections in
`“predicted minus actual" to form section pairs.
`
`' 2
`
`. Fer each pair of sections check if it could be a valid map data issue. Sdfilter out the
`section pair if any of the following happens.
`'
`
`.-
`
` .-
`
`
`
`
`
`B. Both sections too long - difficult to isolate issue and could be driver
`behaviour
`
`
`
`
`wgmwmagmm
` film«2%mu
`
`W5t
`@2121.miii
`
`0. Both Sections of similar length - grobably driver behaviour
`Only‘section pairs for which at least one of the sections is of. reasonable length and the
`difference in lengths are high are retained.
`_
`'
`
`
`
`
`
`Potential issue (retained after the above filters)
`
`3. Return all the lat/longs which are still retained as problematic lat/longs.
`
`Aggregating across trips to find issues
`
`For each trip we now have two sets of problematic lat/longs, those in “actual minus predicted"
`and those in “predicted minus actual”. We then aggregate this across a large number of trips (all
`the trips in a day in a city) and the lat/longs which occur the most number of times in either of
`the two cases are problematic lat/longs.
`
`Since the lat/longs are arbitrary precision points, we geohash them by rounding them off to the
`first 4 decimal places before aggregation. This is to ensure that we consider two lat/longs which
`only differ ”in the 5th decimal as the same point.
`
`
`
`
`
`'
`
`‘
`
`-.
`
`-
`
`-
`
`Output for delhi afier'aggreéation
`a
`
`_
`
`.
`
`Clustering points to find all lat/longs associated with one issue
`
`We then cluster all the lat/longs based on the trips Which affected them. If two lat/longs were
`affected by a lot of common trips then we combine them into the same cluster.
`‘
`We use a common similarity measure between lat/longs known as Jaccard similarity to achieve-
`this.
`.
`
`
`
`
`
`1.0
`
`
`
`MAD Tool Example
`
`Tools and configurations used:
`
`o Tool was run on 100K trips for Bangalore which took place on a single given day.
`
`Methodology: '
`c We have robust ways to filter out the following
`0 Driver behaviour —We filter out the noise caused by driver taking alternative route.
`0 Service roads and main roads - We now filter out the noise where one of
`
`suggested/actual is a service road and other is a main road as they are
`essentially travelling in the same road and it does not actually imply an issue in
`map data.
`0 GPS noise - We have not used map matching as we also want to find missing
`roads in OpenStreetMap (OSM) roads that are not present in OSM. This meant
`that we had to use more robust algorithms to remove GPS Noise (jumpy GPS
`
`points)
`
`Results:
`
`Fig. 3 is the heat map of actual minus predicted.
`Fig. 4 is the heat map of predicted minus actual.
`
`Precision:
`
`We looked at 20 random issues among the top 30-40 issues (based on frequency of occurrence
`of the issue) in both of the heat maps combined. We found the precision to be around 0.9 for
`the sample. Around 90% were actual issues in both the heat maps.
`
`Sample issues of various kinds:
`
`We found a lot of issues of various kinds in the sample analysis. We are documenting one of
`each kind to show the variety of issues our tool detects and the kind of heatmap it produces for
`each of those issues.
`
`11
`
`
`
`4
`
`,
`
`'
`
`'
`
`—
`
`Road segments missing
`
`, I
`
`SUGGESTED AND ACTUAL
`-
`TRACE MINUS PREDICTED
`This missing road issue was presentIn 360 trips in the day. ( out of 100K)-
`Road segments which are extraneous with a corresponding
`miSsing rOad connection
`'
`-
`OSM has anextraneous road connection next to Bellary road. The corresponding straight
`segment in Bellary road is missing. This issue has occurred >2000 times(2%)In the day.
`I
`‘
`-
`
`,o
`
`
`
`
`
`SUGGESTED A_ND ACTUAL ‘
`
`TRACE MINUS PREDICTED
`
`Road segments which ere wrongly connected.
`The road here~is wrongly connected and OSM route suggests you to go straight while all trips‘
`go left take a U-Turnand then come back to the road. This issue has occurred more than 1000
`times.
`
`13
`
`
`
`mm;
`5
`l
`
`,,,,,
`
`§
`,4, g .:~
`
`..
`
`. i
`5
`.
`
`‘\
`
`"EC ,
`i\;~.\\/’ ‘\
`‘-‘
`“L?
`’fl \
`\ 3
`\f
`.
`.
`\\‘
`‘
`
`-
`
`‘
`
`‘ a
`
`“3.,
`”M
`V
`/>
`
`,
`nun/anode
`
`
`
`
`
`.
`»:
`
`. i
`1}
`
`'
`
`_.
`
`‘
`
`‘
`
`3
`
`I
`
`‘
`
`,'
`
`73’
`-i
`‘3, g:
`g
`C 3’
`
`i
`e
`i:
`
`t:
`
`-
`
`7
`?
`
`5:
`i
`i
`‘
`i
`i
`I
`’,
`
`‘
`
`g
`'
`E
`
`
`
`~
`,
`”,1 "
`:v'
`I
`
`5.
`l
`.
`l
`
`
`
`
`
`E ”
`>‘z
`x ’
`(’31
`> i
`'1- '
`g;
`O
`
`N
`
`‘
`
`'
`
`*7
`“
`’
`
`
`,’
`
`‘1'
`.‘
`
`I
`l
`‘
`
`SUGGESTED AND ACTuAL
`
`TRACE MINUS PREDICTED
`
`Missing turn restrictions
`
`The right turn restriction in outer ring road is missing as a result of which all the drivers have to
`go a few 100 meters forward take a U turn come back and then take a left. This is a high traffic
`road which might easily add 5 minutes to the total time. This issue has occurred 500 times.
`
`
`
`TRACE MINUS PREDICTED
`
`‘
`
`SUGGESTED AND ACTUAL
`
`14
`
`
`
`Missing directionality
`
`This road(Magrath road) is missing directionality: All the drivers need to take the detour which is
`significantly longer and has a much higher time to traverse across. This issue has occurred 350
`times.
`'
`
`'
`
`' S
`
`UGGESTED AND ACTUAL
`0
`
`PREDICTEE) MIN‘US TRACE
`
`Wrong road conneCtion across lanes
`
`(
`
`Here there is a wrong read connection from the right lane to the left lane due to which.OSM
`"*
`suggests to go straight and take this connection and go across to the other lane. Since this
`connection does not exist, all the drivers instead have to go in the other direction, take a'U-Turn
`and come all the way back. This issue would add at least 5 minutes to ETA. This issue occurred
`, 320 times in the day.
`’
`'
`'
`‘
`
`‘
`
`r ~
`
`I
`
`
`
`'
`
`OSM shows that there is a roundabout Whereas there is just a normal right turn/left turn present
`in the junction. This has occurred 400 times in the day.
`’
`'
`'
`
`SUGGESTED AND ACTUAL“
`
`Wrong geometry of roads
`s
`
`l6.
`
`u
`
`-
`
`‘
`
`.
`
`.‘
`
`'
`
`-,‘
`
`.
`
`'
`
`‘
`
`,
`
`.
`
`;
`
`
`
`
`
`3T6
`
`SUGGESTED AN ACTJAL
`
`TRACE MINUS PREDICTED
`
`Additional Embodiments:
`
`o The algorithm can prioritise issues that have more effect on ETA/RPQ first rather than
`just the count of the number of times it occurred.
`0 Automatically highlight the differing section in each of the trips for that particular issue
`(similar to the screenshots present in the tables in this doc) and send it to the operators.
`
`l7
`
`
`
`Detection tool for Maps data issues
`
`Introduction
`
`Detection tool’s motivation is to-b\e able to programmatically detect various maps data issues
`using trip traces. Maps issues that it can potentially detect are missing roads, incorrect road
`connections, incorrect geometry, incorrect and missing turn restrictions. This tool is agnostic to
`what the underlying maps data provider is and thus can be used to find maps data issues in any
`‘ of them.
`'
`-
`'
`'
`/
`;
`'
`Current status of this tool is only a proof of conceptand serves as‘a proposal for International
`Maps team to build this in production.
`
`~
`Method
`We compare the actual route taken With the route predicted byrour routing engine based on the
`provider's maps data and look at the difference between them. We account for the density of
`1
`trips as well to be able to have more data points to increase confidence level and also prioritise.
`impact ofucorrécting the flagged issues.
`‘
`For example,
`
`at
`
`
`
`Actual route
`
`Actual route — Provider
`predicted route
`
`Provider predicted
`route - Actual route
`
`'
`
`We then geohash these points in the “diff” between the routes and compute histograms of
`geohashes and look at those geohashes which had a very high count in either of the two \
`~histograms(actual minus Provider or Provider minus actual) and these would help us give a set
`of geohashes(a|l lat longs in a really small region are mapped to the same geohash) which
`could potentially have some issues. These issues could be anything ranging from false positives
`like poor GPS signals at the road leading to mistakes(which we should filter out in the tool) to
`r
`
`18
`
`
`
`wrong directionality on a major road( Provider or OSM might keep predicting the route in the '
`wrong direction but it will never be taken and hence will show Up in our heatmap).
`
`,
`Proof of concept:
`We took 40K trips in Bangalore on a single day and ran those queries through a routing engine
`using OSM data to get route predictions. We then computed histograms as described above.
`We could find the following issues in'OSM data based on the histogramfloutput.
`Fig. 1 is the set of top 500 points which are actually taken by drivers but rarely predicted by
`OSM.
`-
`_
`_
`i
`'
`‘
`»
`-
`‘
`Fig. 2 is the set of top 500 points which are rarely taken by drivers but‘often predicted by QSM.
`Note: Clicking on the points in the histograms gives the lat, long of the point and the UUle of
`trips where the difference occurred.
`.
`'
`'
`
`Here are some sample issueswe found by looking through the above histograms and looking at
`the corresponding trips. Our hypothesis is that a good number'of the followmg issues'can'bé"
`(semi) automated and'brought down to a few 1008 of potential issues which the operators could
`look at. and reject(if it is a false positive) or fix.
`-
`'
`"
`3‘:
`"
`r
`'
`
`I
`o Wrong geometry of‘roads
`u 1 Vidyana
`gfl‘fla. -‘
`
`.
`
`.
`
`f 7 I“
`
`,"
`
`
`
`l9
`
`
`
`I)
`
`Predicted and Taken
`
`.mder...P
`
`a,det
`
`nekaT.dn
`
`Nonexistent turns
`
`.
`
`showing these lat longs in the predicted route but they?-
`Heatmap showed that OSM kept
`I
`were rarely taken. On'inspection we found that this underpass did not exist and OSM was
`wrong.
`
`v.
`
`.1:.
`
`Heatmaproutput of OSM
`
`Aotual
`
`‘%
`' We estimate that arodnd 400 of the 40K trips we looked at had this one.issue(whioh is 1
`
`l).
`
`'
`
`Here are some sample trips._
`;
`
`20
`
`
`
`
`
`
`
`.'
`413:
`
`d L
`a:
`J" v
`In“?
`. [1% »,
`r; 1".
`x.
`..
`
`a
`
`a
`
`..
`
`.,
`y C," “4“
`M ‘ .‘
`'9
`
`I
`
`.1
`
`-
`
`l
`
`'
`
`l
`
`i
`
`I
`
`a
`
`l
`
`'
`
`x.
`
`@
`
`a.
`‘19
`Q5
`
`1‘
`
`“
`
`‘
`
`w
`
`’
`
`\.
`
`"._
`-
`
`‘
`
`
`
`
`“,\\\K‘
`
`~4
`
`\
`
`\A\
`
`‘
`
`1;?
`--
`.
`\u
`,-
`‘3'“;
`\‘K‘
`““3; ‘
`‘.-
`X
`\
`h} ,
`.3
`‘3'».
`\
`aka
`
`SWJcek'rlan)
`YdalMMl
`_
`.
`.
`_‘_.
`"‘w-‘L’; ..
`
`’
`
`.
`
`I»
`‘4) "
`)7
`f. ’3:
`‘.
`g
`31 I
`‘
`» "3'11"; ‘
`,1 }
`E
`‘V ‘1.-‘:~-
`(.11
`K“ \‘i
`:
`H ~
`a f:
`d
`if
`__-1 . Add-.. Va 43‘. A. \“L___.__. _ _-
`_
`
`“madam
`
`‘~-
`
`1
`
`Blue is the predicted route by OSM
`and light red/yellow is the actual
`route taken by the drivers.
`
`Here is another sample we found of incorrect turn restrictions we found in the map.
`
`
`
`.E
`
`M.— ,.,.
`.
`m,
`G“RD ‘
`R;
`
`,R”
`
`Heatma'p(A(;tual - OSM)
`
`Trips 'affected:
`Predicted route in btue
`
`‘9:
`{fa
`
`HéA‘
`fpw
`“W
`
`15.Wu.
`nu.
`
`M
`S,mm.
`
`
`
`
`
`
`OSM route _
`,.
`
`0» Missing directionality(norma| roads)
`
`n‘
`
`Yr” . t
`
`‘
`
`23
`
`
`
` ' Actual route trace
`
`Conclusions
`
`0 The above examples illustrate that, for a lot of map issues, it is possmle to automatea
`lot of the process and identify the issues in the map data.
`'
`'
`‘
`a An advantage of the above‘trip-by-trip" approach of finding map issues (that'Is on 'aper ‘
`trip basis, identifying the differences between actual and suggested routes) is that it
`allows us to find issues in and fix high impactroads (roads with most trips') first.
`0 Another advantage of doing a per trip analysis is that it can drive real-time feedback.
`o
`if a mm is blocked starting today, our system is capable ofIfinding thisIssue fast
`enough to give a feedback as soon as there are a few trips where the turn is.
`_
`predicted by us but not taken
`,
`_
`0 We can use this info to recognize in real time that some roadsarebeingaVOIded
`by a lot of drivers, which can help us detect a trafficam or roadblockageIn reai
`time
`'
`
`
`
`[0001]
`
`Fig. 7 is a high—level block diagram illustrating physical components of a computer
`
`700 used to train or apply computer models such as those including a direct and indirect network
`
`as discussed herein. Illustrated are at least one processor 702 coupled to a chipset 704. Also
`
`coupled to the chipset 704 are a memory 706, a storage device 708, a graphics adapter 712, and a
`
`network adapter 716. A display 718 is coupled to the graphics adapter 712. In one embodiment,
`
`the functionality of the chipset 704 is provided by a memory controller hub 720 and an I/O
`
`controller hub 722. In another embodiment, the memory 706 is coupled directly to the processor
`
`702 instead of the chipset 704.
`
`[0002]
`
`The storage device 708 is any non-transitory computer—readable storage medium,
`
`such as a hard drive, compact disk read-only memory (CD—ROM), DVD, or a solid-state memory
`
`device. The memory 706 holds instructions and data used by the processor 702. The graphics
`
`adapter 712 displays images and other information on the display 718. The network adapter 716
`
`couples the computer 700 to a local or wide area network.
`
`[0003]
`
`As is known in the art, a computer 700 can have different and/or other components
`
`than those shown in FIG. 7. In addition, the computer 700 can lack certain illustrated
`
`components.
`
`In one embodiment, a computer 700, such as a host or smartphone, may lack a
`
`graphics adapter 712, and/or display 7l 8, as well as a keyboard or. external pointing device.
`
`Moreover, the storage device 708 can be local and/or remote from the computer 600 (such as
`
`embodied within a storage area network (SAN)).
`
`[0004]
`
`As is known .in the art, the computer 700 is adapted to execute computer program
`
`modules for providing functionality described herein. As used herein, the term “module” refers
`
`to computer program logic utilized to provide the specified functionality. Thus, a module can be
`
`implemented in hardware, firmware, and/or software.
`
`In one embodiment, program modules are
`
`
`
`stored on the storage device 708, loaded into the memory 706, and executed by the processor
`
`702.
`
`[0005]
`
`The foregoing description of the embodiments of the invention has been presented for
`
`the purpose of illustration; it is not intended to be exhaustive or to limit the invention to the
`
`precise forms disclosed. Persons skilled in the relevant art can appreciate that many
`
`modifications and variations are possible in light of the above disclosure.
`
`[0006]
`
`Some portions of this description describe the embodiments of the invention in terms
`
`of algorithmsand symbolic representations of operations on information. These algorithmic
`
`descriptions and representations are commonly used by those skilled in the data processing arts
`
`to convey the substance of their work effectively to others skilled in the art. These operations,
`
`while described functionally, computationally, or logically, are understood to be implemented by
`
`computer programs or equivalent electrical circuits, microcode, or the like. Furthermore, it has
`
`also proven convenient at times. to refer to these arrangements of operations as modules, without
`
`loss of generality. The described operations and their associated modules may be embodied in
`
`software, firmware, hardware, or any combinations thereof.
`
`[0007]
`
`Any of the steps, operations, or processes described herein may be performed or
`
`implemented with one or more hardware or software modules, alone or in combination with
`
`other devices.
`
`In one embodiment, a software module is implemented with a computer program
`
`product comprising a computer-readable medium containing computer program code, which can
`
`be executed by a computer processor for performing any or all of the steps, operations, or
`
`processes described.
`
`[0008]
`
`Embodiments of the invention may also relate to an apparatus for performing the
`
`operations herein. This apparatus may be specially constructed for the required purposes, and/or
`
`26
`
`
`
`it may comprise a general—purpose computing device selectively activated or reconfigured by a
`
`computer program stored in the computer. Such a computer program may be stored in a
`
`non-transitory, tangible computer readable storage medium, or any type of media suitable for
`
`storing electronic instructions, which may be coupled to a computer system bus. Furthermore,
`
`any computing systems referred to .in the specification may include a single processor or may be
`
`architectures employing multiple processor designs for increased computing capability.
`
`[0009]
`
`Embodiments of the invention may also relate to a product that is produced by a
`
`computing process described herein. Such a product may comprise information resulting from a
`
`computing process, where the information is stored on a non—transitory, tangible computer
`
`readable storage medium and may include any embodiment of a computer program product or
`
`other data combination described herein.
`
`[0010]
`
`Finally, the language used in the specification has been principally selected for
`
`readability and instructional purposes, and it may not have been selected to delineate or
`
`circumscribe the inventive subject matter. It is therefore intended that the scope of the invention
`
`be limited not by this detailed description, but rather by any claims that issue on an application
`
`based hereon. Accordingly, the disclosure of the embodiments of the invention is intended to be
`
`illustrative, but not limiting, of the scope of the invention, which is set forth in the following
`
`Claims.
`
`Date 07 November 2017
`
`To,
`The Controller of Patents
`The Patent Office at Chennai
`
`PRASHANT PHILLIPS
`IN/PA—1229
`
`Agent for the Applicant
`
`27
`
`
`
`Applic_ant(s) Name
`ication No
`App]
`
`UBER TECHNOLOGIES, IN C.
`
`Sheet No.: 1
`5
`Total No. of Sheets
`
`M_.
`
`4/\_,
`
`m.h
`
`
`
`.522.5A:gaféfiz,_
`
`6%a»
`
`yw,.
`
`
`
`#5:,_A,.,My.,,»
`
`,AM
`
`Q,A+A;
`..,,L¢M\,...351::
`
`
`A:1:.,@ufiznwmcmwm_:
`
`.35330max‘a,,..,.,,_A,x_.,«axma..
`
`
`
`mamamémumgm..,,
`
`163¢.«Ava.
`.«525W,
`,l,,as
`
`.,.L.j,.,,,.«Wm,LA,803‘3“:
`ena.
`,G.
`
`.L.Jr,...snaigfiu.M,..
`_N63515.A,1,.__.in...A._..,,.A,7:.Eméé§q
`
`
`
`a.,,,.AM».u./,/
`
`1V.k...f‘.
`9.3.74..,,.mvéy,My_,M.
`
`
`
`
`
`1..5.32.??me
`\E?_
`.A8:32;\
`
`YEW.
`
`_a‘
`
`,.3QNa:rmy
`5.85333;35W,
`
`
`_
`
`.3.2__3AA.A:s.e.,
`
`J.A.yit?i
`
`.$33356,,
`
`
`
`J.,3,(”Eliazfiwaflé
`
`.(wéég.mim.mm,.A..«MAw_w...
`\a.Wa.ww”M.232.3,?Awmm1m.
`,,aM..,w42.32\maA?.m.
`
`.3:.nA«WA.4
`
`
`
`PRA SHANT PHILLIPS
`IN/PA-1229
`
`‘ Agent for the Applicant
`
`
`
`
`
`
`
`
`
`Applicant(§)‘Name: UBER TECHNOLOGIES, INC.’
`Application No:
`‘
`
`‘ Sheet: No.: 2
`Total No. of Sheets: 5 '
`
`“$3913.. “if"
`
`
`
`
`
`
`
` 513.3%?
`
`
`
`
`PRASHAN'T PHILLIPS
`-
`IN/PA-1229
`
`Agent for the Applicant
`
`
`
`Applicant(s) Name: UBER TECHNOLOGIES, INC.
`’
`\
`Application No:
`
`Sheet No.: 3
`
`Total No. of Sheets: 5
`
`
`
`en
`
`u
`
`PRASHANT PHILLIPS
`.1
`a
`IN/PA‘-1229
`
`Agent'for the Applicant
`
`
`
`
`
`Applicanfis) Name: UBER TECHNOLOGIES
`_ Application No:
`
`C
`
`T.
`
`1mt
`
`a%hf
`So
`
`.5u.Mm..H
`
`4.6.5.“End
`
`..0..N_
`m/
`ov
`~I
`\ltf
`
`~
`
`.1
`
`win
`
`_.
`
`4
`.:" ’. 1
`
`,iPRASHANT PHILLIPS
`a
`‘.
`IN/PA-1229
`
`*‘¥'-"i'fi"r€Age-nt for the Applicant
`
`
`
`
`
`Applicant(s) Name: UBER TECHNOLOGIES, INC.
`Application No:
`
`Sheet No.: 5
`Total No. of Sheets: 5
`
`
`
`239
`
`POJ‘NTSNG DEVICE
`135:
`
`
`
`PRASHANT PHILLIPS
`IN/PA—1229
`
`Agent for the Applicant
`
`