You are here

Geophysics Sensors and AI/ML for Subterranean Shipyard Voids and Piles


OUSD (R&E) MODERNIZATION PRIORITY: Artificial Intelligence (AI)/Machine Learning (ML); Autonomy; Nuclear TECHNOLOGY AREA(S): Information Systems; Materials / Processes; Sensors OBJECTIVE: Develop and demonstrate successful seismic geophysical assessment solution to enable non-destructive subterranean assessment of void and pile locations and dimensions (seeking up to 80 feet of penetration) for piers, wharfs, relieving-platforms, and other shipyard-type structures) for initial load restriction or load capacity planning during Port Damage Repair and Port/Harbor/Shipyard assessment scenarios, when electromagnetic contract methods fail due to salt-saturated soils and water. DESCRIPTION: Facility Inspection, Sustainment, and Resilience via Geophysical Assessment Methods, Via Seismic Geophones: Currently, those inspecting waterfront facilities (such as piers, wharfs, relieving-platforms, and other shipyard-type structures) for structural and soil voids or support-pile details cannot assess the subterranean structural components or defects which they cannot see. Also, many geophysical assessment methods, which are applicable inland, are impeded in part or whole by typical waterfront facilities site conditions such as soil types, geology, construction materials, construction configurations, onsite electrical interference, etc. Methods thus eliminated include those which rely on magnetics, electromagnetics, electrical methods, gravity, and nuclear [Ref 1]. Geophysical methods not eliminated include seismic methods [Ref 1]. This SBIR topic is therefore limited to this class of technically feasible methods. The Sensors (Geophones): The geophysical assessment sensors which receive the seismic energy are geophones (hydrophones in waterborne surveys) or commonly referred to as “phones”, and are typically configured for the geological site conditions of the average inland geophysicist rather than for the needs of those working on the waterfront and littoral regions. Therefore, there is a need and room for innovation within the materials, dimensions, and configuration for prototyping specialized geophone devices; and for evaluating within salt-saturated sediments and other structural configurations typical of waterfront and shipyard facilities. Interpretation of Geophone (Seismograph) Data; Improvements via Artificial Intelligence/Machine Learning (AI/ML): The equipment that records input geophone voltages in a timed sequence is the seismograph. In general, the subsurface characterization provided by geophysical exploration methods (to the seismograph data) is valuable for waterfront facilities evaluations for the following reasons: 1. They allow nondestructive investigation below the surface of the ground, pavement, pier deck, or other structure. 2. They allow collection of data over large areas in very much shorter times than most destructive methods. 3. They cost less per data point than most invasive methods. 4. They can offer accurate and timely information for design quality and performance. Although geophysical methods provide the above advantages, it is important to remember that the information obtained in geophysical surveys is often subject to more than one reasonable interpretation. Therefore, there is room for innovation in applying AI/ML to traditional geophysical assessment seismograph data. Combined Need/Opportunity: The needs expressed herein includes improvements in both the prototyping of specialized geophone devices and the accompanying AI/ML software improvements to the seismograph data. The related technical challenges include limited access to real-world facilities and limited (yet available) real-world subterranean defect data. Therefore, there is an opportunity to simulate the subterranean geophysics of the subject scenarios; however, any proposed simulation should be field verified or otherwise calibrated to relevant real-world data. Capability Requirements: Proposals shall address or otherwise exhibit the ability to address the 1.) Specialized design and manufacturing of the requisite specialized sensors, and 2.) The AI/ML aspects of improving (interpreting) seismograph data. Proposals shall also address 3.) The teams experience with: • The typical geophysics and construction of piers, wharfs, relieving-platforms, and other shipyard-type structures; • The design and prototyping of geophysics sensors; and • AI/ML relevant to the subject opportunity. Performance Parameters: This research seeks an overall 30% improvement in a user’s ability to correctly determine subterranean void and pile locations and dimensions, up to a depth of 80 feet of soil penetration. The improvements can come from any combination of improving either the sensors or the AI/ML interpretation of seismograph data or any other aspect of the demonstrated prototype. Note: This SBIR topic does not specify nor limit the innovation of the class of waves, nor the sub-class of waves (i.e., body wave class, surface wave class nor the sub-classes of waves within each of the classes). PHASE I: Determine the technical feasibility of improving and prototyping specialized geophone device(s) for geophysical evaluation within salt-saturated sediments and other structural configurations typical of waterfront and shipyard facilities; for finding: void location(s) and dimensions, subterranean pile location(s) and dimensions, to include driven pile depth; and for proposing the targeted level of improvement in just the geophone sensors. Apply innovative AI/ML to the traditional geophysical assessment seismograph data. Propose the targeted level of improvement in just the AI/ML interpretation/clarification of the seismograph data. Address if improvements are to come from other aspect(s) of the prototype to be demonstrated. Address if and to what extent the AI/ML training data will rely on simulation versus real-world training data. (Note: During the Phase I period of performance, the Navy can make some representative as-built drawings and inspection data available for all subject facility types. The Navy will not provide seismograph data.) Propose how the prototyped sensors will be adapted for underwater use, with a maximum operating depth of 90 feet of seawater (fsw). Propose how the prototyped sensor wave-source will be adapted for underwater use, with a maximum operating depth of 90 fsw. Suggest to what extent the above improvements could reduce the required users’ level of training, the recency of training, and the overall level of experience in order to correctly employ the prototyped device in either routine field application or expeditionary (communications denied) environments. 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: Prototype development of: 1. Specialized Geophysics sensors for use in the salt-saturated soils of waterfront facilities (such as piers, wharfs, relieving-platforms, and other shipyard-type structures), or integration to enable improved data input, when performing data collection via geophones. 2. AI/ML application to automate the clarification and classification of subterranean (seismograph) data for the same site conditions and structures. While not required at this point, possible steps for the above might include: • Development, procurement, and/or manufacture specialized sensors, such as geophones • Gather or simulate relevant AI/ML training data (Government will provide traditional as-built drawings of representative structures, but not seismograph data) • Determining or establishing situ/constructed pattern recognition (while allowing for constructed variability), either via pattern recognition methods, AI/ML, convenient parametric user interface for identification, or other diverse void or pile identification techniques • Locate and classify subterranean void and structure detail, down to UNIFORMAT-II component level [Ref 3], i.e., delineate piles (pile depth), pile-caps, beams, deck, voids (size), etc. • Determining or establishing the construction pattern, while allowing for constructed variability • Conduct field validation of any formerly simulated or approximated training data used in developing the AI/ML neural network • Tabulate or map the prototyped outputs, including voids, piles, and possibly other structural details The Government will provide traditional as-built drawings of representative structures. The Government will also make demonstration facilities available to the Phase II awardee. However, the Phase II awardee will be required to meet all site access requirements; i.e., the Government will not be at fault for the Phase II awardee’s failure to complete the typical site access requirements, either in forms/submittals or in the eligibility of its personnel. The idealized data(s) for structure(s) and defect scenario(s) shall be provided by the awardee, but shall be of typical waterfront and shipyard facilities, and shall include subterranean voids, piles, and other relevant structural details. Single construction type for timber relieving platform is acceptable for Phase II; additionally, conventional concrete pile supported pier is acceptable as a minimum addition. Validation of the following: • Location(s) and dimension(s) of subterranean voids in timber-constructed relieving platform structures • Location(s) and approximate dimension(s) of subterranean piles of timber-constructed relieving platform structures • Location(s) and approximate dimension(s) of other subterranean structural details of timber-constructed relieving platform structures • Constructed structural pattern (i.e., bent/row grid, or similar) • Identification of missing element(s) from pattern or other provision for enhanced user understanding • Increased user correct interpretation of subterranean details by at least 30% overall, compared to current terrestrial geophones and non-AI/ML aided interface, when the same are applied to waterfront and shipyard-type structures • Likelihood that the solution will work by users with low-level training in either routine applications or communications-denied expeditionary applications. Deliver working prototype sensors with integrated elements of the AI/ML application by the end of the full Phase II. PHASE III DUAL USE APPLICATIONS: The expected transition of the product within the Government will include field demonstration of the Phase II solution for one actual timber-constructed relieving platform shipyard wharf/berth (for void location and classification) and one concrete-constructed convention pier (for driven pile depths); where actual gross defects may or may not exist, and where some aspect of the process may be simulation-based, with either simulated or real-world replicated voids, defects, debris, rubble, and/or other realistic anomalies. The Phase III solution will conclude as a Government off the shelf (GOTS) product that the Navy Expeditionary Combat Command (NECC), the Underwater Construction Team (UCT), or the Navy Mobile Construction Battalion (NMCB) may employ during PDR exercises. There is great commercial value in automating the interpretation of seismograph data for waterfront facilities, namely shipyard and port/harbor infrastructure. Therefore, the awardee could transition a non-military tool to industry, possibly in the form of licensing or selling the solution to major vendor(s) of related sensor systems, or computer aided design and modelling tools and software. REFERENCES: 1. Wightman, W et al. “Application of Geophysical Methods to Highway Related Problems.” Report Number: FHWA-IF-04-021, September 1, 2003.; 2. Heffron, Ronald E., ed. “Waterfront Facilities Inspection and Assessment.” ASCE Manuals and Reports on Engineering Practice No. 130. 3. “NAVFAC Design-Build RFP Uniformat Structure.” (UNIFORMAT II / WORK BREAKDOWN STRUCTURE; Section H – Waterfront; see all H1010 through H1040 codes.) 4. “Navy Tactical Reference Publication 4-04.2.9: Expedient Underwater Construction and Repair Techniques.” August 2011.; 5. Unified Facilities Criteria (UFC): 4-150-07 MAINTENANCE AND OPERATION: MAINTENANCE OF WATERFRONT FACILITIES.” June 19, 2001. KEYWORDS: Geophysical; Geophysical method; Geophysical assessment; Geophysical investigation; Geophysical surveys; Geophone; Seismograph; Ultraseismic; Subterranean assessment; Subterranean void; Subterranean pile; Nondestructive testing
US Flag An Official Website of the United States Government