Surface finish quality evaluation system and method

1. An automated drywall finishing system comprising:

a mobile base unit;

a robotic arm that extends between a base end and a distal end, the robotic arm coupled to the mobile base unit at the base end of the robotic arm;

an end effector coupled at the distal end of the robotic arm;

a surface evaluation system comprising at least one light and one or more vision systems that generate target surface data to evaluate quality of a surface of a wall assembly, wherein the one or more vision systems take one or more surface evaluation images or surface evaluation measurements of the surfaces; and

a computing device executing a computational planner that:

obtains the target surface data from the one or more vision systems;

determines based on the target surface data that a portion of the surface of the wall assembly is below a finish quality threshold;

determines a position error of the automated drywall finishing system based at least in part on position data, wherein the position data, corresponding to a previous position of the automated drywall finishing system when the one or more surface evaluation images or surface evaluation measurements were taken, is generated based at least in part on the one or more surface evaluation images or surface evaluation measurements;

generates instructions for driving the end effector, the robotic arm, and the mobile base unit to perform one or more drywalling tasks via the end effector based at least in part on the target surface data, the instructions including tuning parameters of at least one of the end effector, the robotic arm, and the mobile base unit to improve quality of the surface of the wall assembly based at least in part on the target surface data, and wherein the one or more drywalling tasks include finishing the portion of the surface of the wall assembly that is below the finish quality threshold, wherein the finishing improves the portion of the surface to be at or above the finish quality threshold; and

drives, the end effector, the robotic arm, and the mobile base unit to perform the one or more drywalling based on the generated instructions.

2. The automated drywall finishing system of claim 1 , wherein the one or more vision systems comprise one or more of a 3D scanner, laser scanner, and a camera.

3. The automated drywall finishing system of claim 1 , wherein the at least one light is controlled by the surface evaluation system when generating the target surface data, including one or more of:

modulating intensity of the at least one light;

modulating a wavelength of the at least one light;

changing an incident angle of the at least one light; and

turning the at least one light on and off.

4. The automated drywall finishing system of claim 1 , wherein at least one of the at least one light and the one or more vision systems are disposed on the end effector.

5. The automated drywall finishing system of claim 1 , wherein at least one of the at least one light and the one or more vision systems are disposed on the mobile base unit.

6. The automated drywall finishing system of claim 1 ,

wherein the computational planner generate instructions for driving the end effector, the robotic arm, and the mobile base unit to perform the one or more drywalling tasks via the end effector based further on the position error.

7. The automated drywall finishing system of claim 1 , wherein the surface evaluation system configures the at least one light to simulate room lighting conditions.

8. The automated drywall finishing system of claim 1 , wherein the computational planner uses wall assembly information from uploaded at least one of an architectural drawing and a building information model (BIM) to establish the finish quality threshold.

9. The automated drywall finishing system of claim 1 , wherein the surface evaluation system configures the at least one light to apply a harshest lighting setting to the surface to highlight defects in the surface.

10. The automated drywall finishing system of claim 1 , wherein the computational planner uses wall assembly information from uploaded at least one of an architectural drawing and a building information model (BIM) to set lighting generated by the at least one light to be representative of room lighting determined based on the wall assembly information or an environmental lighting model.

11. A method comprising:

obtaining, by a computational planner executing on a computing device of an automated drywall finishing system, target surface data from one or more vision systems of a surface evaluation system of the automated drywall finishing system, wherein:

the automated drywall finishing system comprises the surface evaluation system, a computing device excuting a computational planner, a mobile base unit, a robotic arm, and an end effector,

the robotic arm extends between a base end and a distal end,

the robotic arm is coupled to the mobile base unit at the base end of the robotic arm,

the end effector is coupled at the distal end of the robotic arm, and

the surface evaluation system comprises at least one light and the one or more vision systems that generate the target surface data to evaluate quality of a surface of a wall assembly, and the one or more vision systems take one or more surface evaluation images or surface evaluation measurements of the surface;

determining, by the computational planner, based on the target surface data that a portion of the surface of the wall assembly is below a finish quality threshold;

determining, by the computational planner, a position error of an automated drywall finishing system, based at least in part on position data, wherein the position data, corresponding to a previous position of the automated drywall finishing system when the one or more surface evaluation images or surface evaluation measurements were taken, is generated based at least in part on the one or more surface evaluation images or surface evaluation measurements;

generating, by the computational planner, instructions for driving one or more of the end effector, the robotic arm, and the mobile base unit to perform one or more drywalling tasks based at least in part on the target surface data, the instructions including ming parameters of at least one of the end effector, the robotic arm, and the mobile base unit to improve quality of the surface of the wall assembly based at least in part on the target surface data, and wherein the one or more drywalling tasks includes finishing the portion of the surface of the wall assembly that is below the finish quality threshold, and the finishing improves the portion of the surface to be at or above the finish quality threshold; and

driving, by the computational planner, one or more of the end effector, the robotic arm, and the mobile base unit to perform the one or more drywalling tasks based on the instructions.

12. The method of claim 11 , wherein the surface comprises a coating and wherein the coating is tinted to highlight topography of the coating on the surface, and wherein the target surface data comprises topography data generated by the one or more vision systems detecting the topography.

13. The method of claim 11 , wherein the one or more vision systems comprise a thermal imaging camera, and wherein one or more images generated by the thermal imaging camera are to identify one or more portions of the surface where a coating has been applied to the surface.

14. The method of claim 11 , further comprising using building assembly information uploaded from at least one of a building drawing or a building model to establish the finish quality threshold.

15. The method of claim 11 , further comprising determining a finish quality of the surface based on calibration data and one or more images generated by the one or more vision systems, and

wherein the calibration data comprises at least one of topography measurements of the surface and human labeled datasets.

16. The method of claim 11 , further comprising determining a finish quality of the surface based on two or more surface finish characteristics of the surface, the two or more surface finish characteristics including at least two of:

surface texture,

surface porosity,

size and number of defects on the surface,

surface reflectivity,

surface sheen, and

surface planarity, and

wherein the determining the finish quality of the surface comprises determining whether the two or more surface finish characteristics meet a respective defined finish quality threshold for the two or more surface finish characteristics.

17. The method of claim 11 , wherein the one or more vision systems comprise a thermal imaging camera, and wherein one or more images generated by the thermal imaging camera are to identify areas of different coating thicknesses applied to the surface.

18. The method of claim 11 , wherein the one or more vision systems comprise a thermal imaging camera, and wherein one or more images generated by the thermal imaging camera are to determine that coating applied to the surface has dried and meets criteria to be sanded.

19. The method of claim 11 , further comprising using building assembly information uploaded from at least one of a building drawing or a building model to set lighting generated by at least one light to be representative of room lighting determined based on the building assembly information or an environmental lighting model.

20. The method of claim 11 , further comprising:

controlling the at least one light by the surface evaluation system when generating the target surface data, wherein the controlling includes at least one of:

modulating intensity of the at least one light;

modulating a wavelength of the at least one light; and

changing an incident angle of the at least one light.