N213-140 TITLE: Automated Pier Battle Damage Assessment from 3D Scanned Data
OUSD (R&E) MODERNIZATION PRIORITY: Artificial Intelligence (AI)/Machine Learning (ML); Autonomy
TECHNOLOGY AREA(S): Information Systems; Materials / Processes; Sensors
OBJECTIVE: Enable an automated pier repair planning tool with the inputs from remote sensing-captured infrastructure data such as three-dimensional (3D) point-cloud, Simultaneous Localization and Mapping (SLAM) technologies, photogrammetry, and Structure from Motion (SfM), and the outputs being Battle Damage Assessment (BDA) Rough Order of Magnitude (ROM) for repair type, enabling material quantities, and estimated repair times.
DESCRIPTION: Currently, when Port Damage Repair (PDR) efforts are conducted for piers or wharves, geo-referenced 3D point-cloud data for a structure is gathered via LiDAR (for the above water structure) and multi-beam sonar (for the underwater structure). However, presently, all key details of the scanned data are manually entered into a spreadsheet-based tool, known as the Pier Reconnaissance Assessment Tool (PRAT), for facility repair planning and detailed repair instructions. This manual data entry is a laborious human-in-the-loop bottleneck and is an opportunity for significant PDR improvement. Therefore, automating the conversion of structural 3D scan data into actionable tabular gross-defects and BDA are the key focus of this SBIR topic. The methods employed to achieve this end are believed to have commercial value.
This SBIR topic seeks to prototype the automation of gross-defect and battle damage detection (from structure scan 3D point-clouds, SLAM technologies, photogrammetry, and SfM data types), defect identification, defect volume approximation, defect location, and defect tabular summation. This SBIR topic must enable future tabulation of ROM for repair material quantities, and ROM for approximated repair times. All ROM estimates may be nominally approximated from conventional construction, as some military-specific solutions are still in development and such military-specific information is considered out of scope for proposers to this SBIR topic. Common pier construction types to be considered include cast reinforced concrete, pre-stressed concrete, steel, and timber construction (listed in the order of importance, with emphasis on cast reinforced concrete construction). In addition to the concrete, steel, or timber sub-structure or base-structure concerns, the facilities’ fender (bumper) system and mooring hardware (cleats and bollards) shall also be addressed.
This SBIR topic does not address generation of repair instruction, plans, specifications, etc., as the actual repair methods may or may not be of a conventional construction method.
Current methods for converting 3D point-cloud data into Building Information Modeling (BIM), or for inventorying of scanned city streets, as applied to waterfront structures fall short since they rely on libraries of standardized pre-modelled mechanical components. However, with the construction of piers and wharves, while there are common construction techniques and configurations per material type, there is notable variability within even a single structure, i.e., piers are not built with uniformity, precision, or accuracy (particularly in regard to pile placement and angle, pile-cap dimensions, cast deck features, etc.). Therefore, innovation is needed to post-process 3D scan data, delivering volumetric construction details and patterns on the existing and missing component(s), while allowing for original structural variability (i.e., variability is not a gross defect or battle damage).
Also, current methods for defect detection/location rely on change-detection between two vintages of data. However, in the subject case, the user is assumed to not have access to pre-event scan data. Therefore, gross-defect and BDA will need to rely on things such as in-situ pattern recognition/missing-pattern detection within a +/- 12 inch grid precision, and +/- 6 inch feature/component (e.g., pile, pile cap, etc.) dimensional precision (statistical pattern configuration [i.e., change-detection strategy] from undamaged portions of structure), innovations in artificial intelligence (AI)/machine learning (ML), a convenient user interface for identification, or other diverse BDA techniques. Increases in the level of required human interaction for this step will proportionally lower the overall satisfaction in the resulting solutions(s).
It is desired to reduce the time (or labor equivalent) required between obtaining of scan data to the completion of the BDA tabular data entry by a factor of between half (satisfactory) and three quarters (excellent) reduction.
This SBIR topic seeks solutions that will work equally well for structure scan data sets from either (listed in order of preference): 1.) 3D point-clouds, 2.) SLAM technologies, 3.) Photogrammetry, and 4.) SfM technologies. Proposed solutions that do not address all these listed technologies will receive proportionally less consideration. Emphasis for this SBIR topic is currently placed on, however not limited to, 3D point-cloud data.
This SBIR topic seeks solutions which can be executed in the field, without reliable Wi-Fi connectivity; therefore, are not cloud-based or require high computing capability. This topic also seeks solutions that utilize open standard data interfaces and enables interoperability between IT systems.
Once the gross defects and BDA are tabulated, with ROM repair volumes and times summarized, the requirements of this topic will be satisfied.
PHASE I: Determine the technical feasibility of automating the conversion of structural 3D scan data into actionable tabular-based gross-defects and BDA. Within this requirement, separately determine the technical feasibility of:
a) Post-processing 3D scan data, delivering volumetric detail and construction patterns on the existing and potentially missing component(s), while allowing for constructed variability.
b) Determining BDA from in-situ pattern recognition, missing pattern detection (i.e., 3D statistically-based pattern-detection/change-detection based on undamaged portions of a variably-constructed structure), innovations in AI/ML, a field user interface for identification, or other diverse BDA techniques.
c) Reducing by half or three quarters the time (or labor equivalent) between obtaining scan data through to the completion of the BDA tabular data entry. For proposal purposes, assume a concrete-constructed pier approximately 100 ft. wide x 1,000 ft. long x 5 ft. of average under-deck clearance, with 100 bents and 20 piles per bent (i.e., 20 rows of piles); assume that a three person BDA assessment team will require two days (equivalent to 48 labor hours) to assess.
d) The solution’s likelihood to work with 3D point-clouds, SLAM technologies, photogrammetry, and SfM data types, in a communications degraded or communications denied environment (i.e., local connectivity possible, global/networked connectivity not).
Note: Beginning with commercial off-the-shelf (COTS) options is acceptable in Phase I. Limited proof of concept for custom integration is also acceptable in Phase I, but is not required.
PHASE II: Develop a prototype of custom solutions or integration that enables post-processing of 3D scan data of an idealized structure(s) and idealized damage scenario(s). Deliver a tabular summary of volumetric detail, location, and affected structural components (down to NAVFAC Design-Build RFP Structure [UNIFORMAT-II] component level) for gross-defects and for BDA.
While not required at this point, possible steps for the above might include:
• Determining or establishing situ/constructed pattern recognition (while allowing for constructed variability), either via pattern recognition methods, AI/ML, field user interface for identification, or other diverse defect identification techniques
• Providing volumetric detail of the structure, down to UNIFORMAT-II component level (see references), i.e., delineate the volume of each pile, pile-cap, beam, deck span, etc.
• Determining or establishing construction pattern for the missing component(s), while allowing for constructed variability.
• Providing volumetric detail of the missing structure component(s), down to UNIFORMAT-II component level, i.e., enabling future ROM repairs and times likely driven by the combination of volume and component location.
• Providing tabular output of volumetric detail, location, and affected structural component for gross-defects and for BDA.
Provide the idealized data(s) for structure(s) and damage scenario(s) of typical port/harbor pier(s) and wharf construction types, and include rubble, debris, and other simulated realistic scenario for the solution to overcome. (Note: Single construction type for reinforced concrete is acceptable for Phase II.)
Provide validation of the following:
• Volumes of constructed element(s)
• Constructed structural pattern (i.e., bent/row grid, or similar)
• Volumes of missing/damaged element(s)
• Identification of missing element(s) from pattern or convenient graphical user interface
• Reductions by half to three quarters for the time (or labor equivalent) between obtaining scan data through to the completion of the BDA tabular data generation
• The solution’s likelihood to work with 3D point-clouds, SLAM technologies, photogrammetry, and SfM data types
• Capability to operate in a communications degraded or communications denied environment (i.e., local connectivity possible, global/networked connectivity not)
PHASE III DUAL USE APPLICATIONS: Transition the product within the Government to include field demonstration of the Phase III solution for two actual concrete-constructed piers, where actual gross defects may or may not exist, and where the actual data is edited to simulate battle damage with simulated debris, rubble, and other realistic anomalies.
Revise the tabular formatting of the Phase II solution to fully satisfy employment by the Pier Reconnaissance Assessment Tool (PRAT) process.
Potential dual-use applications include:
1) Government off-the-shelf (GOTS) to U.S. Army Corps of Engineers (USACE), Engineer Research and Development Center (ERDC) for use with the PRAT; whereby Navy Expeditionary Combat Command (NECC) and the Underwater Construction Team (UCT) will employ the solution from within the PRAT.
2) A non-military tool for licensing or selling to major vendor(s) of related computer aided design and modelling tools and software.
1. Navy Tactics, Techniques and Procedures NTTP 4-04.2.9 Expedient Underwater Construction and Repair Techniques.
2. Unified Facilities Criteria (UFC); UFC 4-150-07; MAINTENANCE AND OPERATION: MAINTENANCE OF WATERFRONT FACILITIES.
3. UFC 4-150-08; INSPECTION OF MOORING HARDWARE.
4. NAVFAC Design-Build RFP Uniformat Structure; UNIFORMAT II / WORK BREAKDOWN STRUCTURE; Section H – Waterfront; see all H1010 through H1040 codes.
5. ASCE Manuals and Reports on Engineering Practice No. 130; Waterfront Facilities Inspection and Assessment.
KEYWORDS: Battle Damage Assessment; BDA; Engineering Survey; Expeditionary Pier Repair; Repair Planning Tool; 3D; Point Cloud Data; Point Cloud Conversion; Simultaneous Localization and Mapping; (SLAM); Photogrammetry; Structure From Motion; Building Information Modeling; BIM
** TOPIC NOTICE **
The Navy Topic above is an "unofficial" copy from the overall DoD 21.3 SBIR BAA. Please see the official DoD Topic website at rt.cto.mil/rtl-small-business-resources/sbir-sttr/ for any updates.
The DoD issued its 21.3 SBIR BAA pre-release on August 25, 2021, which opens to receive proposals on September 21, 2021, and closes October 21, 2021 (12:00pm edt).
Direct Contact with Topic Authors: During the pre-release period (August 25 thru September 20, 2021) proposing firms have an opportunity to directly contact the Technical Point of Contact (TPOC) to ask technical questions about the specific BAA topic. Once DoD begins accepting proposals on September 21, 2021 no further direct contact between proposers and topic authors is allowed unless the Topic Author is responding to a question submitted during the Pre-release period.
Topic Q&A System: After the pre-release period, proposers may submit written questions through the DoD Topic Q&A System at www.dodsbirsttr.mil/topics-app/. In the Topic Q&A System, the questioner and respondent remain anonymous but all questions and answers are posted for general viewing. Topic Q&A will close to new questions on October 7, 2021 at 12:00 p.m. ET
Note: Questions should be limited to specific information related to improving the understanding of a particular topic’s requirements. Proposing firms may not ask for advice or guidance on solution approach and you may not submit additional material to the topic author. If information provided during an exchange with the topic author is deemed necessary for proposal preparation, that information will be made available to all parties through the system.
Topics Search Engine: Visit the DoD Topic Search Tool at www.dodsbirsttr.mil/topics-app/ to find topics by keyword across all DoD Components participating in this BAA.