Laser waveform embedding

A method comprising:

• receiving, by a processor, from a light detection and ranging (LIDAR) device, and for each respective point of a plurality of points in an environment, a corresponding waveform that represents physical characteristics of the respective point;
• determining, by the processor and based on the corresponding waveform of each respective point, a map of the environment that includes a representation of a corresponding position of the respective point within the environment;
• determining, by the processor, for each respective point, and based on processing the corresponding waveform of the respective point by a machine learning (ML) model of an embedding model, a corresponding embedding vector, wherein the corresponding embedding vector comprises a plurality of values that represent the physical characteristics of the respective point in a vector space defined by the embedding model;
• and
• detecting or classifying, by the processor, a physical feature within the environment based on (i) the corresponding positions of one or more points of the plurality of points and (ii) the corresponding embedding vectors of the one or more points.

A non-transitory computer readable storage medium having stored thereon instructions that, when executed by a computing system, cause the computing system to perform operations comprising:

• receiving, from a light detection and ranging (LIDAR) device and for each respective point of a plurality of points in an environment, a corresponding waveform that represents physical characteristics of the respective point;
• determining, based on the corresponding waveform of each respective point, a map of the environment that includes a representation of a corresponding position of the respective point within the environment;
• determining, for each respective point and based on processing the corresponding waveform of the respective point by a machine learning (ML) model of an embedding model, a corresponding embedding vector, wherein the corresponding embedding vector comprises a plurality of values that represent the physical characteristics of the respective point in a vector space defined by the embedding model;
• and
• detecting or classifying a physical feature within the environment based on (i) the corresponding positions of one or more points of the plurality of points and (ii) the corresponding embedding vectors of the one or more points.