Techniques for real-time mapping in a movable object environment

Abstract

Techniques are disclosed for real-time mapping in a movable object environment. A real-time mapping system can include at least one movable object including a computing device, a scanning sensor electronically coupled to the computing device, and a positioning sensor electronically coupled to the computing device. The computing device can include at least one processor and a mapping manager, the mapping manager may be configured to obtain mapping data from the scanning sensor and obtain positioning data from the positioning sensor. The mapping manager can associate the mapping data with the positioning data based at least on time data associated with the mapping data and the positioning data, and then generate a map in a first coordinate system based at least on the associated mapping data and positioning data.

What is claimed is: 
     
         1 . A system for real-time mapping in a movable object environment, comprising:
 at least one movable object including a computing device;   a scanning sensor electronically coupled to the computing device;   a positioning sensor electronically coupled to the computing device;   the computing device including at least one processor and a mapping manager, the mapping manager including instructions which, when executed by the processor, cause the mapping manager to:
 obtain mapping data from the scanning sensor; 
 obtain positioning data from the positioning sensor; 
 associate the mapping data with the positioning data based at least on time data associated with the mapping data and the positioning data; and 
 generate a map in a first coordinate system based at least on the associated mapping data and positioning data, the map comprising a plurality of voxels and wherein an occupancy probability threshold value is used to determine whether a voxel includes a point in the map. 
   
     
     
         2 . The system of  claim 1 , wherein the instructions to associate the mapping data with the positioning data based at least on the time data associated with the mapping data and the positioning data, when executed by the processor, further cause the mapping manager to:
 upsample the positioning data to include a number of positions equal to a number of points in the mapping data; and   associate each point in the mapping data to a corresponding position in the upsampled positioning data.   
     
     
         3 . The system of  claim 2 , further comprising:
 clock circuitry electronically coupled to the scanning sensor and the positioning sensor, to provide the time data associated with the mapping data and the positioning data.   
     
     
         4 . The system of  claim 1 , wherein the instructions to generate the map comprising the plurality of voxels in the first coordinate system based at least on the associated mapping data and positioning data, when executed by the processor, further cause the mapping manager to:
 for each voxel of the plurality of voxels of the map:
 determine one or more points from the mapping data to be located in the voxel; and 
 determine an occupancy probability for the voxel based at least on a number of points in that voxel. 
   
     
     
         5 . The system of  claim 4 , wherein the occupancy probability is determined based on a variance of the positioning data associated with the one or more points located in the voxel. 
     
     
         6 . The system of  claim 4 , wherein the instructions, when executed by the processor, further cause the mapping manager to:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculate an average position of the one or more points located in the voxel; and 
 generate a point in the map at the average position. 
   
     
     
         7 . The system of  claim 6 , wherein the instructions, when executed, further cause the mapping manager to:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculate an average intensity value of the one or more points located in the voxel; and 
 associate the average intensity value with the generated point in the map. 
   
     
     
         8 . A method for real-time mapping in a movable object environment, comprising:
 obtaining mapping data from a scanning sensor supported by a movable object;   obtaining positioning data from a positioning sensor supported by the movable object;   associating the mapping data with the positioning data based at least on time data associated with the mapping data and the positioning data; and   generating a map in a first coordinate system based at least on the associated mapping data and positioning data, the map comprising a plurality of voxels and wherein an occupancy probability threshold value is used to determine whether a voxel includes a point in the map.   
     
     
         9 . The method of  claim 8 , wherein generating the map comprising the plurality of voxels in the first coordinate system based at least on the associated mapping data and positioning data, further comprises:
 for each voxel of the plurality of voxels of the map:
 determining one or more points from the mapping data to be located in the voxel; and 
 determining an occupancy probability for the voxel based at least on a number of points in that voxel. 
   
     
     
         10 . The method of  claim 9 , wherein the occupancy probability is determined based on a variance of the positioning data associated with the one or more points located in the voxel. 
     
     
         11 . The method of  claim 9 , further comprising:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculating an average position of the one or more points in the voxel; and 
 generating a point in the map at the average position. 
   
     
     
         12 . The method of  claim 11 , further comprising:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculating an average intensity value of the one or more points in the voxel; and 
 associating the average intensity value with the generated point in the map. 
   
     
     
         13 . The method of  claim 8 , further comprising:
 determining a distribution of points in the mapping data, each point in the mapping data associated with a distance from a nearest neighboring point in the mapping data; and   removing any points associated with a distance greater than a distance threshold value.   
     
     
         14 . The method of  claim 8 , further comprising:
 downsampling the mapping data by a scaling factor;   dividing the mapping data into the plurality of voxels; and   outputting an average point from the downsampled mapping data for each of the plurality of voxels.   
     
     
         15 . A non-transitory computer readable storage medium including instructions stored thereon which, when executed by one or more processors, cause the one or more processors to:
 obtain mapping data from a scanning sensor supported by a movable object;   obtain positioning data from a positioning sensor supported by the movable object;   associate the mapping data with the positioning data based at least on time data associated with the mapping data and the positioning data; and   generate a map in a first coordinate system based at least on the associated mapping data and positioning data, the map comprising a plurality of voxels and wherein an occupancy probability threshold value is used to determine whether a voxel includes a point in the map.   
     
     
         16 . The non-transitory computer readable storage medium of  claim 15 , wherein the instructions to generate the map comprising the plurality of voxels in the first coordinate system based at least on the associated mapping data and positioning data, when executed, further cause the one or more processors to:
 for each voxel of the plurality of voxels of the map:
 determine one or more points from the mapping data to be located in the voxel; and 
 determine an occupancy probability for the voxel based at least on a number of points in that voxel. 
   
     
     
         17 . The non-transitory computer readable storage medium of  claim 16 , wherein the instructions, when executed, further cause the one or more processors to:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculate an average position of the one or more points located in the voxel; and 
 generate a point in the map at the average position. 
   
     
     
         18 . The non-transitory computer readable storage medium of  claim 17 , wherein the instructions, when executed, further cause the one or more processors to:
 for each voxel having the occupancy probability greater than the occupancy probability threshold value:
 calculate an average intensity value of the one or more points in the voxel; and 
 associate the average intensity value with the generated point in the map. 
   
     
     
         19 . The non-transitory computer readable storage medium of  claim 15 , wherein the instructions, when executed, further cause the one or more processors to:
 transform the map into a second coordinate system; and   output the transformed map.   
     
     
         20 . The non-transitory computer readable storage medium of  claim 15 , wherein the scanning sensor includes a light detection and ranging (LiDAR) sensor, and wherein the positioning sensor includes a real time kinematic (RTK) sensor or an inertial measurement unit (IMU) sensor.