Search

US-12623286-B2 - Methods and apparatus for mobile additive manufacturing of advanced roadway systems

US12623286B2US 12623286 B2US12623286 B2US 12623286B2US-12623286-B2

Abstract

The present disclosure provides various aspects for mobile and automated processing utilizing additive manufacturing. The present disclosure includes methods processing vision system data. Vision system image data may be processed by artificial intelligence algorithms. The artificial intelligence algorithms may perform GAN based synthesis of image edits to plan deposits of sized features upon detected defects of a surface. An artificial intelligence chip may perform rapid image processing on the fly as a mobile additive manufacturing apparatus functions.

Inventors

  • Robert A. Flitsch
  • Frederick A. Flitsch

Assignees

  • Robert A. Flitsch
  • Frederick A. Flitsch

Dates

Publication Date
20260512
Application Date
20240606

Claims (20)

  1. 1 . A method for depositing materials upon a surface, the method comprising: receiving a first image representation in a controller of an apparatus from a vision system of the apparatus, wherein the apparatus is a mobile additive manufacturing apparatus comprising: a controller capable of executing algorithms and providing control signals, an additive manufacturing system to deposit a line feature material at prescribed locations across a surface of a roadway according to a first digital model processed by the controller, a drive system operative to transport the additive manufacturing system along the surface, a vision system to receive images of the surface of the roadway, a navigation system to determine a location of the mobile additive manufacturing system and guide the drive system, and a power system capable of providing power to operate at least the drive system, navigation system, controller and additive manufacturing system; recognizing a registration mark printed upon the surface, and comparing a size and a location of the registration mark to a model and modifying the first digital model to generate control signals for the additive manufacturing system of the additive manufacturing apparatus; transmitting a control signal to the additive manufacturing system of the additive manufacturing apparatus, adding a first material to the surface with the mobile additive manufacturing apparatus and wherein the first material is added to the surface in a first pattern; moving the mobile additive manufacturing apparatus with the drive system to a new location; confirming the location of the mobile additive manufacturing system with the navigation system; and adding the first material to the surface in a second pattern according to the first digital model when the mobile additive manufacturing system is at the new location.
  2. 2 . The method of claim 1 further comprising analyzing the first image representation with an artificial intelligence algorithm, wherein the analyzing of the first image representation with the artificial intelligence algorithm extracts features and semantic segmentation aspects from the image.
  3. 3 . The method of claim 2 wherein the artificial intelligence algorithm has been trained to utilize images and associated feature and semantic segmentation extraction to perform a first editing of the first image representation to create a second image representation.
  4. 4 . The method of claim 3 wherein the artificial intelligence system has been trained to apply a predetermined edit to a feature type.
  5. 5 . The method of claim 4 wherein the predetermined edit of the feature type is used to create an editing vector which is imbedded into the latent space of the artificial algorithm.
  6. 6 . The method of claim 5 wherein the second image representation is created by running the artificial intelligence algorithm with the latent space resulting from applying the editing vectors.
  7. 7 . The method of claim 6 wherein the predetermined edit of the feature type performs at least a change of the feature's two dimensional size.
  8. 8 . The method of claim 6 wherein the predetermined edit of the feature type performs at least a change of the feature's two dimensional location.
  9. 9 . The method of claim 6 wherein the predetermined edit of the feature type performs at least a change of the feature's three dimensional height, wherein the image comprises a three dimensional representation of the surface.
  10. 10 . The method of claim 3 further comprising receiving a second image representation from the vision system of the mobile additive manufacturing apparatus, wherein the second image representation shows at least in part a view of the processed area of the surface.
  11. 11 . The method of claim 10 further comprising analyzing the second image representation with the artificial intelligence algorithm and comparing features in the second image representation with corresponding features in the first image representation to determine whether an adjustment of parameters of the editing vectors is required.
  12. 12 . A method for depositing materials upon a surface, the method comprising: receiving a first image in a controller of an apparatus from a vision system of the apparatus, wherein the apparatus is a mobile additive manufacturing apparatus comprising: a controller capable of executing algorithms and providing control signals, an additive manufacturing system to deposit a line feature material at prescribed locations across a surface of a roadway according to a first digital model processed by the controller, a drive system operative to transport the additive manufacturing system along the surface, a vision system to receive images of the surface of the roadway, a navigation system to determine a location of the mobile additive manufacturing system and guide the drive system, and a power system capable of providing power to operate at least the drive system, navigation system, controller and additive manufacturing system; analyzing the first image representation with an artificial intelligence algorithm, wherein the analyzing of the first image representation with the artificial intelligence algorithm extracts features and semantic segmentation aspects from the image; transmitting a control signal to the additive manufacturing system of the additive manufacturing apparatus, adding a first material to the surface with the mobile additive manufacturing apparatus and wherein the first material is added to the surface in a first pattern; moving the mobile additive manufacturing apparatus with the drive system to a new location; confirming the location of the mobile additive manufacturing system with the navigation system; and adding the first material to the surface in a second pattern according to the first digital model when the mobile additive manufacturing system is at the new location.
  13. 13 . The method of claim 12 wherein the artificial intelligence algorithm has been trained to utilize images and associated feature and semantic segmentation extraction to perform a first editing of the first image representation to create a second image representation.
  14. 14 . The method of claim 13 wherein the artificial intelligence system has been trained to apply a predetermined edit to a feature type.
  15. 15 . The method of claim 14 wherein the predetermined edit of the feature type is used to create an editing vector which is imbedded into the latent space of the artificial algorithm.
  16. 16 . The method of claim 15 wherein the second image representation is created by running the artificial intelligence algorithm with the latent space resulting from applying the editing vectors.
  17. 17 . The method of claim 16 wherein the predetermined edit of the feature type performs at least a change of the feature's two dimensional size.
  18. 18 . The method of claim 16 wherein the predetermined edit of the feature type performs at least a change of the feature's two dimensional location.
  19. 19 . The method of claim 16 wherein the predetermined edit of the feature type performs at least a change of the feature's three dimensional height, wherein the image comprises a three dimensional representation of the surface.
  20. 20 . The method of claim 12 wherein the analyzing the first image representation with an artificial intelligence algorithm comprises at least in part utilizing an artificial intelligence chip to perform the artificial intelligence algorithm processing steps.

Description

CROSS REFERENCE TO RELATED APPLICATION This application claims priority to the U.S. Non-Provisional patent application Ser. No. 18/209,907 filed Jun. 14, 2023 as a continuation in part application, which in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 17/728,291 filed Apr. 25, 2022 as a continuation application, which in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 17/072,029, filed on Oct. 15, 2020 as a Continuation in Part application, which in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 16/853,046, filed on Apr. 20, 2020 as a Continuation in Part Applications, which in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 16/692,993, filed on Nov. 22, 2019 as a Continuation Application, which in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 16/233,439, filed on Dec. 27, 2018 as a Continuation Application. The application Ser. No. 16/233,439 in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 15/963,767, filed on Apr. 26, 2018 as a Continuation Application. The application Ser. No. 15/963,767 in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 15/641,509, filed on Jul. 5, 2017 as a Divisional Application. The application Ser. No. 15/641,509 in turn claims priority to the U.S. Non-Provisional patent application Ser. No. 14/310,443, filed on Jun. 20, 2014 as a Continuation in Part. The application Ser. No. 14/310,443 in turn claims the benefit of the U.S. Provisional Application Ser. No. 61/838,302 filed on Jun. 23, 2013. The application Ser. No. 15/561,509 also claims priority to the U.S. Non-Provisional patent application Ser. No. 14/310,556, filed on Jun. 20, 2014 as a Continuation in Part. The U.S. Non-Provisional patent application Ser. No. 16/692,993, filed on Nov. 22, 2019 claims priority to the U.S. Non-Provisional patent application Ser. No. 15/639,766, filed on Jun. 30, 2017 as a Continuation in Part. The contents of each are hereby incorporated by reference. FIELD OF THE INVENTION The present disclosure relates to methods and apparatus that support mobile additive material processing. Robotic and human controlled mobility may be combined with additive manufacturing techniques that “print” or additively deliver materials to specific locations over significant distances. The methods and apparatus may be applied to the productions of advanced building structures and roadways. BACKGROUND OF THE INVENTION A known class of approaches to material fabrication can be classified as additive manufacturing. Material in various forms, including solid, powder, gel, gas or liquid forms may be processed in such a manner to deposit or lock in material in a target location in space. Numerous techniques may be utilized to perform additive manufacturing. In extrusion processes, materials in wire or filament form are controlled by an extrusion head which may be moved above a work area. The use of multiple extrusion heads and extrusion material may allow for both permanent and temporary structures to be formed. By building the extruded material in layers or in regions, complex shapes may be formed in three dimensions. However, the technology is limited by the dimensions of the work space—the ability of the head or heads to move in the two dimensions of a plane and also by the dimension of the ability of the head to move vertically relative to a planar support structure. There may be numerous variations on this form of additive manufacturing. A different class of additive manufacturing may be classified as Stereolithography. In this class, a light or heat source is used to transform the material in space. In some Stereolithography implementations, the work support plane is submerged in a photoactive or thermo-active liquid and a laser or other light or heat source is rastered across a thin surface layer of the liquid between the support structure and the top level of the liquid. By translating the support structure down a layer into the liquid, the fluent nature of the liquid reforms a thin layer of new unreacted material over the work surface or the previously processed layer. Versions of Stereolithography may also work with powder formed starting material. The powder may be shaped into a thin layer and then spatially defined. Lasers may be used to transform portions of the layer into a solidified material. In other examples, other energy sources such as, for example, electron beams, may be used to transform the powder. Various materials including metals, insulators and plastics may be formed into three dimensional shapes by these processing techniques. A different type of processing occurs when a print head is used to deposit material onto the powder. The deposit may chemically react with the powder or may be an adhesive that consolidates the powder into an adhered location. The prevalence of high resolution print