Computer-implemented method and a data processing hardware for processing sensor data points

A computer-implemented method for processing sensor data points, comprising by data processing hardware:

• 1.1. initializing, by data processing hardware, a distributions buffer G for storing an initial plurality of Gaussian distributions gn, each Gaussian distribution g comprising an initial plurality of sensor data points Pr received sequentially from at least one sensor, each Gaussian distribution g having an associated predetermined distribution distance threshold Ddmin;
• 1.2. sequentially acquiring, by the data processing hardware, a new sensor data point Prnew from the at least one sensor and computing a distribution distance Dd between the new sensor data point Prnew and each Gaussian distribution g;
• 1.3. verifying, by the data processing hardware, for each Gaussian distribution g if the distribution distance Dd fulfills a distribution distance condition Dd<Ddmin, respectively distribution distance
• Dd<the associated predetermined distribution distance threshold Ddmin;
• 1.3.1. if no Gaussian distribution g fulfills the distribution distance condition Dd<Ddmin, generating, by the data processing hardware, a new Gaussian distribution gnew, including the new sensor data point Prnew,
• 1.3.2. if the distribution distance condition Dd<Ddmin is fulfilled by a single Gaussian distribution g1, updating, by the data processing hardware, the single Gaussian distribution g1 by adding the new sensor data point Prnew, generating an updated single Gaussian distribution g1-updated,
• 1.3.3. if the distribution distance condition Dd<Ddmin is fulfilled by at least two Gaussian distributions g1, g2, . . . gm, merging, by the data processing hardware, the at least two Gaussian distributions g1, g2 . . . gm, generating a merged Gaussian distribution gm-merged, and updating, by the data processing hardware, the merged Gaussian distribution gm-merged by adding the new sensor data point Prnew, generating an updated merged Gaussian distribution gm-merged-updated,
• 1.4. updating, by the data processing hardware, the initial plurality of Gaussian distributions gn by including the new Gaussian distribution gnew, or the updated single Gaussian distribution g1-updated, or the updated merged Gaussian distribution gm-merged-updated, and
• generating, by the data processing hardware, an updated plurality of Gaussian distributions gn-updated in the distributions buffer G,
• wherein the acts 1.2., 1.3., and 1.4. are repeated for a number n of new sensor data points Prn-new.