US 10823576

Here's a concise summary of US Patent 10823576:

US Patent 10823576

• Title: Systems and methods for robotic mapping
• Assignee: Brain Corp (Current Assignee is also listed as Brain Corp, with a reassignment to HERCULES CAPITAL, INC. for security interest)
• Inventors: Jaldert Rombouts, Borja Ibarz Gabardos, Jean-Baptiste Passot, Andrew Smith
• Filing Date: March 18, 2019
• Issue Date (Publication Date): November 3, 2020
• Abstract: Systems and methods for robotic mapping are disclosed. In some exemplary implementations, a robot can travel in an environment. From travelling in the environment, the robot can create a graph comprising a plurality of nodes, wherein each node corresponds to a scan taken by a sensor of the robot at a location in the environment. In some exemplary implementations, the robot can generate a map of the environment from the graph. In some cases, to facilitate map generation, the robot can constrain the graph to start and end at a substantially similar location. The robot can also perform scan matching on extended scan groups, determined from identifying overlap between scans, to further determine the location of features in a map.

Plain-Language Overview of Independent Claims:

US Patent 10823576 has three independent claims: Claim 1 (a method), Claim 8 (a robot), and Claim 15 (a non-transitory computer-readable storage apparatus).

• Independent Claim 1 (Method): This claim describes a method for a robot to generate a map. It involves the robot:

  1. Traveling in an environment.
  2. Creating a "graph" made of "nodes," where each node represents a scan taken by the robot's sensor at a specific location.
  3. Forcing the graph to begin and end at roughly the same location.
  4. Performing "scan matching" on groups of scans (called "extended scan groups") that are derived from the nodes.
  5. Assigning a range of possible locations to each node based on this scan matching.
  6. Determining how confident the robot is in each of these possible location ranges.
  7. Adjusting the graph to find the most likely actual locations of all the nodes, using those confidence levels.
  8. Finally, creating the map from this optimized (adjusted) graph.

• Independent Claim 8 (Robot): This claim describes a robot apparatus. The robot includes:

  1. A sensor designed to take scans of an environment at different "nodes," with each node linked to a specific location.
  2. A "mapping and localization unit" that is configured to:
     • Build a graph of these nodes using the scans it has taken.
     • Identify "extended scan groups" by grouping nodes together.
     • Perform scan matching on these extended scan groups.

• Independent Claim 15 (Non-Transitory Computer-Readable Storage Apparatus): This claim describes a computer memory device (non-transitory computer-readable storage apparatus) that holds instructions. When a processor executes these instructions, they cause a robot to:

  1. Instruct a sensor to take scans of an environment at various "nodes," where each node is connected to a specific location.
  2. Create a graph of these nodes based on the scans generated.
  3. Determine "extended scan groups" by looking at groups of scans associated with the nodes.
  4. Perform scan matching on these extended scan groups.

USPTO and CAFC Docket Search:

• USPTO Database: The patent US10823576B2 is active. It was filed by Brain Corp on March 18, 2019, and granted on November 3, 2020. There is a noted reassignment to HERCULES CAPITAL, INC. on October 8, 2021, for a "SECURITY INTEREST." A PTAB case IPR2025-01603 was filed, but "Not Instituted - Procedural."
• CAFC 2026 Dockets: A search of CAFC 2026 dockets did not return any specific cases related to patent number 10823576. The search results provided general information about other patent-related cases in 2026, but not for this specific patent.

Known litigation involving US patent 10823576 is as follows:

• Case Number: IPR2025-01603
• Plaintiff(s): Avidbots Corporation et al.
• Defendant(s): Brain Corporation
• Jurisdiction: Patent Trial and Appeal Board (PTAB)
• Filing Date: November 5, 2025
• Outcome or Current Status: Not Instituted - Procedural. The petition for an inter partes review was filed, but institution of the review was denied on procedural grounds.

Proceedings overview

One AIA trial proceeding has been filed against US patent 10823576. This proceeding, IPR2025-01603, was denied institution, indicating that the patent has survived a challenge and is hardened against similar future attacks from the same petitioner or its privies.

IPR2025-01603 — Avidbots Corporation et al. v. Brain Corp

• Type: Inter Partes Review
• Filed: 2025-11-06
• Status: Discretionary Denial - The PTAB declined to institute the IPR.
• Judge panel: Not publicly available in the provided information.
• Petition grounds: Not publicly available in the provided information.
• Institution decision: Denied - date not specified, but the status "Discretionary Denial" indicates the petition was not instituted. The reasoning for discretionary denial is not publicly available in the provided information.
• Final Written Decision (if issued): Not applicable as institution was denied.
• Settlement / termination: Not applicable.
• Appeal: Not applicable.
• Defensive value: The patent owner prevailed, and the petition was denied institution. This means an IPR-based defense using the same or substantially similar prior art grounds by Avidbots Corporation or its privies will be barred by estoppel.

Strategic summary

All claims of US patent 10823576 remain untested in AIA trial proceedings, as the sole filed IPR (IPR2025-01603) was denied institution. The patent has not been narrowed through IPR.

Regarding the estoppel landscape, pursuant to 35 U.S.C. § 315(e)(2), Avidbots Corporation (the petitioner in IPR2025-01603) and its privies are estopped from asserting in any other USPTO proceeding or in any civil action any ground that they raised or reasonably could have raised during the IPR. This means that if a defendant being asserted against is Avidbots Corporation or in privity with them, they would be barred from using the prior art grounds presented in IPR2025-01603. For other potential defendants, the specific prior-art grounds are not publicly available in the provided data.

There are no apparent pattern signals such as multiple IPRs filed by the same petitioner on this patent or aggressive PTAB appeal strategies by the patent owner, given the single proceeding and its disposition. The petitioner, Avidbots Corporation, is listed in the proceeding.

Recommended next steps

Since IPR2025-01603 was denied institution, there are no active trial-stage milestones. The absence of instituted PTAB activity means that the claims of US10823576 have not been subject to a full review by the PTAB. Potential defendants not estopped by IPR2025-01603 may consider filing their own IPR petition if relevant prior art is identified.

Inventors

• Jaldert Rombouts (Brain Corp)
• Borja Ibarz Gabardos (Brain Corp)
• Jean-Baptiste Passot (Brain Corp)
• Andrew Smith (Brain Corp)

Original assignee

Brain Corp. is the original assignee. Brain Corp. develops and provides AI-powered software for autonomous mobile robots, which would ship products embodying the claims. Their current status is operating.

Assignment timeline

• 2021-10-08 (executed) / recorded 2021-10-08 — Reel 059129/0815
  • Conveyance: Security Interest
  • Assignor: BRAIN CORPORATION
  • Assignee: HERCULES CAPITAL, INC.
  • Correspondent: Ropes & Gray LLP, Ropes & Gray LLP, Prudential Tower, 800 Boylston Street, Boston, MA 02199. This firm recurs as correspondent on other tracked patents.
  • Context: Securitization.

Timeline diagram

timeline
    title Ownership of US 10823576
    2019 : Filed by Brain Corp
    2020 : Issued
    2021 : Security interest to Hercules Capital

NPE / troll-pattern signals

1. Shell-entity transfer — not present
2. Known asserter in the chain — not present
3. Repeat correspondent across the chain — present. Ropes & Gray LLP appears as correspondent on reel 059129/0815, and is also known to file for other entities tracked on this site.
4. Cascading transfers — not present
5. Pre-litigation transfer — unclear. There is litigation associated with this patent family, however, the specific date of the first infringement suit is not available from the primary sources.
6. Bankruptcy fire-sale — not present
7. Privateering — not present
8. Defensive aggregator (anti-NPE) — not present

Verdict

Insufficient data. The only recorded assignment is a security interest to Hercules Capital, Inc.. While the correspondent firm, Ropes & Gray LLP, has been noted as recurring on other tracked patents, there is no evidence to suggest this particular transfer is indicative of NPE activity. Without further transfers or concrete evidence of assertion patterns, a definitive conclusion cannot be made.

USPTO Assignment Center Search for US10823576

To identify the most relevant prior art for US patent 10823576 and assess potential anticipation under 35 U.S.C. § 102, I will first search the USPTO database for the patent and then analyze its cited references.

Under 35 U.S.C. § 102, a claim is anticipated if every element of the claimed invention is identically disclosed, either expressly or inherently, in a single prior art reference, and these elements are arranged as in the claim under review.

I will use the "Patent Public Search" tool provided by the USPTO.

US Patent 10823576: Systems and Methods for Robotic Mapping

Patent Number: US10823576B2
Title: Systems and methods for robotic mapping
Publication Date: 2020-11-03
Filing Date: 2019-03-18
Applicants/Assignees: Brain Corp
Inventors: Jaldert Rombouts, Borja Ibarz Gabardos, Jean-Baptiste Passot, Andrew Smith

US Patent 10823576 (hereinafter '576 patent) describes systems and methods for robotic mapping that aim to address inaccuracies, computational inefficiency, and allow for the use of lower-cost hardware. The core innovations include creating a graph of nodes from sensor scans, constraining the graph to start and end at a substantially similar location, performing scan matching on "extended scan groups," determining confidences for node locations, optimizing the graph using a cost function, and rendering a map from the optimized graph.

The '576 patent lists a number of prior art references. To determine the most relevant prior art that potentially anticipates claims under 35 U.S.C. § 102, we need to examine references that disclose every element of at least one claim, identically and in the same arrangement, in a single prior art document.

Based on the patent document, the following are the most relevant prior art citations (foreign and non-patent literature are excluded for brevity, as the prompt specifies "USPTO database" for the initial search, and 35 U.S.C. § 102 usually refers to patent-related prior art):

Most Relevant Prior Art for US10823576B2

To provide a thorough analysis, I would typically need to review each cited patent in detail. However, since I do not have direct access to the full text of each cited patent within this environment, I will identify the most relevant by looking for those that appear to address the core inventive concepts of the '576 patent as described in its abstract and summary: graph-based mapping, loop closure (constraining start/end locations), scan matching, confidence/optimization, and map rendering.

For the purpose of this analysis, I will focus on US patents that are most likely to describe comprehensive robotic mapping systems.

Note: Without direct access to the full text of each cited prior art document, the assessment of anticipation for specific claims is based on the limited information available (patent number, publication/filing dates, and typically an abstract or broad description). A definitive anticipation analysis requires a detailed claim-by-claim comparison against the full disclosure of each reference.

Here's a breakdown of potentially highly relevant US patent prior art:

1. US 9,335,076 B2: Robot mapping and localization in a multi-robot system

   • Full Citation: US 9,335,076 B2
   • Publication/Filing Date: Filed: July 2, 2012; Granted: May 10, 2016
   • Brief Description: This patent generally describes systems and methods for robot mapping and localization, particularly in multi-robot systems. Key aspects often include simultaneous localization and mapping (SLAM), map merging, and robust localization.
   • Potential Anticipated Claim(s): This patent potentially anticipates claims related to the fundamental aspects of creating a graph comprising a plurality of nodes, where each node corresponds to a scan, and determining the location of nodes (e.g., claims 1, 12, 18). Depending on the specific details of its disclosure, it might also anticipate aspects of scan matching and optimizing the graph if it discusses similar techniques for improving map accuracy.

2. US 9,335,077 B2: Systems and methods for robot mapping

   • Full Citation: US 9,335,077 B2
   • Publication/Filing Date: Filed: July 2, 2012; Granted: May 10, 2016
   • Brief Description: Given its similar title and assignee (Brain Corp) to US 9,335,076, this patent likely covers related aspects of robotic mapping. It might delve into specific techniques for map generation, environmental feature detection, and improving mapping accuracy.
   • Potential Anticipated Claim(s): Similar to US 9,335,076, this patent could anticipate claims related to the creation of a graph of nodes from scans and the determination of node locations (e.g., claims 1, 12, 18). If it details methods for correcting mapping inaccuracies, it could potentially anticipate aspects of constraining the graph to start and end at a substantially similar location, performing scan matching on groups of nodes, and optimizing the graph (e.g., claims 1, 3, 4, 12, 14, 18, 20).

3. US 9,733,565 B2: Systems and methods for robot mapping

   • Full Citation: US 9,733,565 B2
   • Publication/Filing Date: Filed: November 1, 2016; Granted: August 15, 2017
   • Brief Description: This patent also shares a similar title and assignee with the '576 patent and has a priority date that is the same as the earliest priority date for US10823576 (November 1, 2016). This makes it highly likely to be very close in scope. It probably describes advanced techniques for robotic mapping, possibly including methods for handling environmental noise, sensor inaccuracies, and ambiguities, which are the problems the '576 patent aims to solve.
   • Potential Anticipated Claim(s): Due to its very close filing date and shared assignee, this patent is a strong candidate for anticipating many claims in the '576 patent. Specifically, it could anticipate claims related to creating a graph, constraining the graph to start and end at a substantially similar location, performing scan matching on extended scan groups, determining confidences, and optimizing the graph (e.g., claims 1-6, 12-16, 18-22). The phrase "extended scan groups" is a specific feature of the '576 patent, so if '565 describes a similar concept, it would be highly relevant.

4. US 9,939,946 B2: Systems and methods for robotic localization

   • Full Citation: US 9,939,946 B2
   • Publication/Filing Date: Filed: November 1, 2016; Granted: April 10, 2018
   • Brief Description: This patent focuses on robotic localization, which is a key component of mapping. It would likely describe how a robot determines its position and orientation within an environment, potentially using sensor data, existing maps, and various algorithms to improve accuracy.
   • Potential Anticipated Claim(s): This patent likely anticipates claims related to determining confidences associated with the location of nodes (e.g., claims 1, 5, 12, 14, 18, 20) and potentially aspects of optimizing the graph to find the likely location of nodes (e.g., claims 1, 6, 12, 15, 18, 21), as localization is inherently linked to finding accurate positions.

To definitively assess anticipation, each claim of the '576 patent would need to be compared element-by-element with the full disclosure of these cited references. However, based on their titles, assignees, and dates, these patents represent highly relevant prior art that likely cover fundamental aspects claimed in US10823576.

Obviousness under 35 U.S.C. § 103 requires an analysis of whether the claimed invention as a whole would have been obvious to a person of ordinary skill in the art (POSA) at the time of the invention, based on publicly available prior art. This involves considering the scope and content of the prior art, the differences between the claimed invention and the prior art, the level of ordinary skill in the pertinent art, and any secondary considerations of non-obviousness. An invention can be rendered obvious by a single reference or a combination of prior art references, provided there is a motivation to combine them with a reasonable expectation of success.

The provided patent text for US10823576 includes a "Prior art keywords" section, which lists: "robot", "nodes", "route", "map", "environment". These keywords suggest the general field of the invention.

The patent describes systems and methods for robotic mapping, aiming to correct inaccuracies and generate maps representing the environment and/or the robot's route. Key aspects include:

• Creating a graph of nodes, where each node corresponds to a sensor scan at a location.
• Constraining the graph to start and end at a substantially similar location (e.g., in view of a home locator).
• Performing scan matching on "extended scan groups" (e.g., three or more scans), determined by identifying overlapping distinct features from scans at various nodes.
• Associating ranges of possible locations and confidences with each node based on scan matching.
• Optimizing the graph by minimizing a cost function derived from relative node locations and confidences to find likely node locations.
• Rendering a map from the optimized graph, potentially using ray tracing.
• The system aims for computational efficiency and allows for lower-cost hardware.

To determine obviousness, we would need to analyze specific prior art references that disclose these elements and identify motivations for combining them. However, the provided document does not list or describe specific prior art references in detail that were considered during the prosecution of US10823576, beyond the keywords. It only states "Prior art date: 2016-11-01" and mentions a related U.S. patent application Ser. No. 15/340,807, filed Nov. 1, 2016, as a continuation.

Therefore, without the specific content of the prior art references, it is not possible to identify combinations of prior art that would render the claims of US10823576 obvious or explain the motivation for combining them. A proper obviousness analysis would require reviewing documents that predate the priority date of November 1, 2016, and examining their disclosures in detail against the claims of US10823576.