Methods and systems for calibrating multiple lidar sensors

A computer-implemented method of executing an Iterative Closest Point (ICP) algorithm on digital datasets in a multidimensional space, the digital datasets comprising:

• (i) a first dataset having a plurality of first data points, and
• (ii) a second dataset having a plurality of second data points and a plurality of normal vectors,
• a given second data point from the plurality of second data points being associated with a respective normal vector from the plurality of normal vectors,
• the method executable by an electronic device, the method comprising, during execution of the ICP algorithm:
• determining, by the electronic device, a plurality of pairs of data points based on at least some of the plurality of first data points and at least some of the plurality of second data points,
• a given one of the plurality of pairs including (i) a respective first data point and (ii) a respective second data point;
• determining, by the electronic device, pair-specific error values for respective ones of the plurality of pairs,
• a given pair-specific error value being indicative of a distance measurement for the respective one of the plurality of pairs in the multidimensional space,
• weighting, by the electronic device, the pair-specific error values by respective weighting factors, thereby determining weighted pair-specific error values,
• a given weighting factor for a respective one of the plurality of pairs being inversely proportional to a density of occurrence of the normal vector of the second data point in the respective one of the plurality of pairs in a distribution of the plurality of normal vectors, such that:
• (i) the more the plurality of normal vectors has vectors that are similar to the normal vector of the second data point in the respective one of the plurality of pairs,
• the more the weighted pair-specific error value of the respective one of the plurality of pairs is reduced in comparison to the respective pair-specific error value, and
• (i) the less the plurality of normal vectors has vectors that are similar to the normal vector of the second data point in the respective one of the plurality of pairs,
• the more the weighted pair-specific error value of the respective one of the plurality of pairs is increased in comparison to the respective pair-specific error value;
• and
• determining, by the electronic device, a global error value for the second dataset as a sum of the weighted pair-specific error values, the global error value being a measurement of a pre-determined error metric that is to be minimized during the execution of the ICP algorithm.

An electronic device comprising:

• a processor;
• a non-transitory computer-readable medium comprising instructions, the instructions for executing at least an Iterative Closest Point (ICP) algorithm;
• the non-transitory computer-readable medium storing at least:
• (i) a first dataset having a plurality of first data points, and
• (ii) a second dataset having a plurality of second data points and a plurality of normal vectors,
• a given second data point from the plurality of second data points being associated with a respective normal vector from the plurality of normal vectors,
• a communication interface for communicating with a sensor mounted on a vehicle,
• the processor, upon executing the instructions, being configured to:
• determine a plurality of pairs of data points based on at least some of the plurality of first data points and at least some of the plurality of second data points,
• a given one of the plurality of pairs including (i) a respective first data point and (ii) a respective second data point;
• determine a pair-specific error values for respective ones of the plurality of pairs,
• a given pair-specific error value being indicative of a distance measurement for the respective one of the plurality of pairs in the multidimensional space,
• weight the pair-specific error values by respective weighting factors, thereby determining weighted pair-specific error values,
• a given weighting factor for a respective one of the plurality of pairs being inversely proportional to a density of occurrence of the normal vector of the second data point in the respective one of the plurality of pairs in a distribution of the plurality of normal vectors,
• determine a global error value for the second dataset as a sum of the weighted pair-specific error values,
• the global error value being a measurement of a pre-determined error metric that is to be minimized during the execution of the ICP algorithm.