N244-D04 TITLE: Direct to Phase II: Next-generation Autonomy for Unmanned Maritime Vehicles (UMVs)

OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Trusted AI and Autonomy

The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws.

OBJECTIVE: Develop next-generation autonomy that will increase the capability and scope of utility of Unmanned Maritime Vehicles (UMVs), while decreasing the level of remote human operator involvement.

DESCRIPTION: The Navy’s Intelligent Autonomous Systems Science and Technology Strategy states the key importance of smaller, more numerous, and distributed capabilities as a complement to larger, more powerful, and proven capabilities, intelligent machines that can adapt in unstructured environments at machine speed given overwhelming data and are attritable when necessary, and decision superiority achieved through ubiquitous and persistent data collection, situational understanding at the edge and support to a robust information warfare capability.1 This SBIR Topic is focused on increasing the capability of Unmanned Maritime Vehicle (UMV) autonomy and increasing the scope of utility of UMVs by enabling the UMV autonomy to be used in a much greater number of situations than it can be today for a much greater variety of tasks. For the purposes of this topic, the term UMV refers to both Unmanned Surface Vehicles (USVs) and Unmanned Underwater Vehicles (UUVs). UMVs are constrained by very low bandwidth and intermittent communications including extended periods of time with no communication with the remote human operator, and substantial limitations in situational awareness. This necessitates highly capable autonomous control software on these platforms that can make decisions without the involvement of a human operator, using incomplete, uncertain and sometimes inaccurate information. Today, mature autonomy software is being fielded on UMVs (for the purposes of this Topic, mature autonomy is defined as that which has been demonstrated in a relevant environment, technology readiness level (TRL) 6) [Ref 2]. Advancements are needed in autonomy’s ability to handle complex tasks and dynamic, unstructured, uncertain and harsh maritime environments without the assistance of a human operator. Autonomy must be able to perform more complex tasks than currently possible, for example, to maneuver around another vessel, or multiple vessels, that are uncooperative and to counter another vessel, or multiple vessels, that are maneuvering in an adversarial manner (maneuvering to counter the UMV’s maneuvers or to threaten the UMV, such as herding or blocking tactics). UMV autonomy needs to support the ability of UMVs to cooperate with each other to work as a team to perform complex tasks, such as optimizing the performance of a distributed sensor network. Technologies needed to realize these advancements are assessed currently to be TRL 1-4. This Topic seeks to raise the TRL level of these technologies.

Next-generation autonomous control for UMVs will increase the capability and scope of utility of UMVs by enabling the UMV autonomy to be used in a much greater number of situations than it can be today, with high degrees of resilience and reliability. The desired future autonomy for UMVs will have the following four capabilities:

(1) ability to perform complex tasks with little to no human intervention,

(2) ability to handle dynamic, unstructured, uncertain and harsh maritime environments,

(3) can support a variety of diverse missions and tasks,

(4) support many heterogeneous UMVs that cooperate autonomously on their perception and decision-making, with the UMVs in that group performing a variety of different tasks.

The autonomy should have the four capabilities listed above while operating with the following two constraints:

(1) low-bandwidth and intermittent communication with a human operator including extended periods of time with no communication between the UMV and human operator, and

(2) be robust to uncertain, inaccurate and intermittent perception information.

Approaches to multi-UMV autonomy should require modest computing resources aboard the UMVs. For example, in biology, control and sensing in schooling fish [Ref 4] or flocking birds enables intricate maneuvers of the group. Biologically inspired approaches to multi-UMV autonomy are in scope of this Topic, but not required; there are many other approaches that can also be envisioned. Future UMVs will encounter dynamic situations in which, for example, other vessels are maneuvering in unexpected ways or conditions are changing and their autonomous control will require the capability to effectively handle the situation with no contemporaneous human operator assistance. The Government anticipates entry TRL levels of 1-4 for this project. Testing in simulation and via small-to-medium physical scale-model testing in-water will be important components of this effort, both to further develop the proposed approach to next-generation autonomy, characterize its performance and establish progress in increasing the TRL level. UMV autonomy should be portable so that it can operate on various UMVs with minimal integration effort and cost. In performing the work under this Topic, the small business concern should be aware of the Unmanned Maritime Autonomy Architecture (UMAA), comply with it where possible, but not be constrained by it if deviations are necessary to achieve the goals of this Topic [Ref 3]. Deviations from UMAA should be documented and rationale for these deviations, in the context of supporting the Topic objectives, should also be documented.

PHASE I: For a Direct to Phase II topic, the Government expects that the small business would have accomplished the following in a Phase I-type effort and developed a concept for a workable prototype or design to address, at a minimum, the basic requirements of the stated objective above. The below actions would be required in order to satisfy the requirements of Phase I:

• Proposals must show that the Offeror understands the current state of the art in UMV autonomy and explain how the proposed approach will advance the state of the art.

• Proposals must describe in detail the Offeror’s concept for maritime autonomy for UMVs. The proposal should clearly explain the rationale for the selection of the proposed concept for next-generation autonomy and how it will satisfy the four capabilities of this topic within the two constraints stated in the Description section above. This rationale must be clearly supported by, for example, analysis, testing in simulation, and/or small scale-model testing. Approaches to next-generation autonomous control that are adapted from systems other than UMVs are of interest and approaches that leverage previous research in this area are also of interest.

• The proposal should describe the approach to testing of the next-generation autonomy algorithms. The proposal must provide a clear explanation of the feasibility of the proposed testing methodology.

FEASIBILITY DOCUMENTATION: Offerors interested in participating in Direct to Phase II must include in their response to this topic Phase I feasibility documentation that substantiates the scientific and technical merit and Phase I feasibility described in Phase I above has been met (i.e., the small business must have performed Phase I-type research and development related to the topic NOT solely based on work performed under prior or ongoing federally funded SBIR/STTR work) and describe the potential commercialization applications. The documentation provided must validate that the proposer has completed development of technology as stated in Phase I above. Documentation should include all relevant information including, but not limited to technical reports, test data, prototype designs/models, and performance goals/results. Work submitted within the feasibility documentation must have been substantially performed by the offeror and/or the principal investigator (PI). Read and follow all of the DON SBIR 24.4 Direct to Phase II Broad Agency Announcement (BAA) Instructions. Phase I proposals will NOT be accepted for this topic.

PHASE II: Using results from the Phase I-type effort, the Phase II effort will develop, demonstrate and validate prototype Next-generation Autonomy software for UMVs that addresses the requirements and constraints stated in the Description section above. The Phase II proposal shall address, at a minimum:

1) Development of algorithms and software capable of addressing the four requirements provided in the Description section of this Topic, while able to operate within the two constraints provided in that section.

2) The Government anticipates that testing to support development of the Next-generation Autonomy software will consist of, at a minimum, simulation-based testing and in-water testing with physical-small scale models. Testing with full scale UMVs is costly and time-consuming and is not a requirement for a successful proposal.

3) Validation of Next-generation Autonomy software will also be done in simulation and with small-to-medium scale physical models in water. The validation phase of the project will consist of structured testing against metrics that the small business will develop as part of this project. The validation phase will elucidate the exit TRL of the Next-generation Autonomy software resulting from this project.

4) Deliverables shall include the Next-generation Autonomy software, a report containing robust documentation of the software (including algorithms, architecture (areas of compliance and non-compliance with UMAA), interfaces, build instructions, necessary software components and environment to build the Next-generation Autonomy software and a software user manual) data acquired during this project, and test methodology and metrics and test results.

5) Composition of the proposed team: teams that are structured to facilitate knowledge transfer of previous research results to this project, for example a small business-university team, while not required, are strongly encouraged.

6) Describe how the software is architected to address cyber security issues and the approach for doing so.

Approaches to perception and communications are not in scope of this Topic, although assumptions should be stated regarding perception and communications capability on the UMV.

The Phase II period of performance is anticipated to be four years; 24 month Base, 24 month Option, if exercised.

PHASE III DUAL USE APPLICATIONS: Given successful completion of the Phase II project and subject to availability of funding, the expected transition of Next-generation Autonomy software resulting from this Phase II SBIR is to the acquisition program for use within its UMAA architecture, in its USV and UUV programs. To enable successful commercialization, Phase III is expected to address integration, via interfaces defined in UMAA, of the Next-generation Autonomy software product of this Phase II SBIR into a larger software architecture, cyber security compliance of the Next-generation Autonomy software, as well as additional rigorous testing of the Next-generation Autonomy software in higher fidelity environments.

REFERENCES:

1. Duffie, Warren, Jr., INNOVATING IAS: DON RELEASES STRATEGY FOR ACCELERATING AUTONOMY, AI DEVELOPMENT, Office of Naval Research, September 15, 2021 https://www.nre.navy.mil/media-center/news-releases/innovating-ias-don-releases-strategy-accelerating-autonomy-ai
2. Department of the Navy Unmanned Campaign Framework, https://nwcfoundation.org/wp-content/uploads/2021/03/20210315-Unmanned-Campaign_Final_LowRes.pdf
3. The Association for Uncrewed Vehicle Systems International (AUVSI) Unmanned Maritime Autonomy Architecture (UMAA) https://www.auvsi.org/unmanned-maritime-autonomy-architecture
4. V. Sridhar, I. Couzin et al., The geometry of decision-making in individuals and collectives, Proceedings of the National Academy of Sciences Vol. 118 No. 50 e2102157118, 2021; https://www.pnas.org/doi/pdf/10.1073/pnas.2102157118

KEYWORDS: Autonomy; Artificial Intelligence; Unmanned Maritime Vehicles; Unmanned Surface Vehicles, Unmanned Underwater Vehicles, Intelligent Autonomous Systems

---