Production of digital road maps by crowdsourcing

1. A computer system for updating and/or supplementing a digital road map through crowdsourcing, the computer system comprising a digital road map and:

a. a plurality of road vehicles traveling on a road network during at least one driving session, each road vehicle comprising a position sensor for measuring a plurality of geographical coordinates of the road vehicle traveling on the road network,

b. at least one data collection server for receiving geographical coordinate measurements associated with each driving session, the geographical coordinate measurements serving as geographical traces,

c. at least one processor for:

i. extracting, for each geographical trace, a trajectory curve passing substantially through all of the measurements of the geographical trace,

ii. detecting inflection points of each trajectory curve, the inflection points serving as vertices,

iii. grouping together all of the vertices into a plurality of vertex classes, using an unsupervised classification algorithm,

iv. selecting a most central vertex in each vertex class, the most central vertex serving as a representative,

v. forming, from each geographical trace, a road segment between representatives that successively intersect the course of the geographical trace when they are considered in pairs,

vi. joining, based on a superposition of the geographical traces, the road segments that successively intersect the course of the superposition of the geographical traces to obtain digital road sections,

vii. identifying redundant road segments based on a distance between the road segments,

viii. deleting the redundant road segments so as to keep only one road segment between two consecutive representatives, and

ix. updating and/or supplementing the digital road map based on the obtained digital road sections.

2. The computer system as claimed in claim 1 , wherein the processor is further configured so as, before updating and/or supplementing the digital road map, to:

a. calculate, for each road segment, a regression function from the geographical traces substantially following the road segment, and

b. adapt the shape of each road segment based on the associated regression function, such that the road segment matches the shape of the regression function.

3. The computer system as claimed in claim 2 , wherein the processor is further configured so as, when calculating the regression function, to:

a. calculate, for each road segment, a measure of statistical dispersion between the geographical traces, wherein the measure of dispersion represents the number of lanes of the road segment.

4. A method for updating and/or supplementing a digital road map through crowdsourcing, the method comprising:

a. providing a plurality of road vehicles for traveling on a road network during at least one driving session, each road vehicle comprising a position sensor for measuring a plurality of geographical coordinates of the road vehicle traveling on the road network,

b. providing at least one data collection server for receiving geographical coordinate measurements associated with each driving session, the geographical coordinate measurements serving as geographical traces,

c. extracting, for each geographical trace, a trajectory curve passing substantially through all of the measurements of the geographical trace,

d. detecting inflection points of each trajectory curve, the inflection points serving as vertices,

e. grouping together all of the vertices into a plurality of vertex classes, using an unsupervised classification algorithm,

f. selecting a most central vertex in each vertex class, the most central vertex serving as a representative,

g. forming, from each geographical trace, a road segment between representatives that successively intersect the course of the geographical trace when they are considered in pairs,

h. joining, based on a superposition of the geographical traces, the road segments that intersect the course of the superposition of the geographical traces to obtain digital road sections,

i. identifying redundant road segments based on a distance between the road segments,

j. deleting the redundant road segments so as to keep only one road segment between two consecutive representatives, and

k. updating and/or supplementing the digital road map based on the obtained digital road sections.

5. The method as claimed in claim 4 , further comprising, before updating and/or supplementing the digital road map:

a. calculating, for each road segment, a regression function from the geographical traces substantially following the road segment, and

b. adapting the shape of each road segment based on the associated regression function, such that the road segment matches the shape of the regression function.

6. The method as claimed in claim 5 , further comprising, during the calculating the regression function:

a. calculating, for each road segment, a measure of statistical dispersion between the geographical traces, wherein the measure of dispersion represents the number of lanes of the road segment.

7. A non-transitory computer program with a program code for executing the steps of a method as claimed in claim 4 when the computer program is loaded into the computer or run on the computer.