HR001118S0027.pdf

Broad Agency Announcement Persistent Aquatic Living Sensors (PALS)

BIOLOGICAL TECHNOLOGIES OFFICE

HR001118S0027

February 15, 2018

HR001118S0027, PALS

TABLE OF CONTENTS

PART I: OVERVIEW INFORMATION

PART II: FULL TEXT OF ANNOUNCEMENT

1. Funding Opportunity Description

1.1. Program Overview

1.2. Program Metrics

1.3. Program Milestones and Deliverables

1.4. Program Demonstrations

1.5. General Requirements

2. Award Information

2.1. General Award information

2.2. Fundamental Research

3. Eligibility Information

3.1. Eligible Applicants

3.2. Organizational Conflicts of Interest

3.3. Cost Sharing/Matching

4. Application and Submission Information

4.1. Address to Request Application Package

4.2. Content and Form of Application Submission

4.3. Funding Restrictions

4.4. Other Submission Requirements

5. Application Review Information

5.1. Evaluation Criteria

5.2. Review of Proposals

5.3. Phase 1 Down-selection Process

6. Award Administration Information

6.1. Selection Notices

6.2. Administrative and National Policy Requirements

6.3. Reporting

6.4. Electronic Systems

7. Agency Contacts

8. Other Information

9. APPENDIX 1 – Volume II Checklist

10. APPENDIX 2 – Platform Space, Weight, Power, and Communications

11. APPENDIX 3 – Sensitivity and Specificity Definition

PART I: OVERVIEW INFORMATION

Federal Agency Name – Defense Advanced Research Projects Agency (DARPA), Biological Technologies Office

Funding Opportunity Title – Persistent Aquatic Living Sensors (PALS) Announcement Type – Initial announcement Funding Opportunity Number – HR001118S0027 Catalog of Federal Domestic Assistance Numbers (CFDA) – 12.910 Research and Technology Development Dates o Posting Date – February 15, 2018 o Proposal Abstract Due Date and Time - March 19, 2018 4 PM ET o Proposal Due Date and Time - April 30, 2018 4 PM ET o Proposers Day – March 2, 2018 https://www.fbo.gov/spg/ODA/DARPA/CMO/DARPA-SN-18-29/listing.html

Concise description of the funding opportunity – DARPA seeks innovative proposals for new systems that employ natural or engineered marine organisms as sensor elements to amplify signals related to the presence, movement, and classification of manned or unmanned underwater vehicles (M/UUV). The Persistent Aquatic Living Sensors (PALS) program will leverage or develop living organisms as sensor transducers, and pair them with a detector and analysis suite to produce deployable sensor systems able to provide timely information on vehicle activity across a wide variety of maritime environments.

Anticipated individual awards - Multiple awards are anticipated.

Types of instruments that may be awarded - Procurement contract, cooperative agreement or other transaction.

Agency contact o Points of Contact The BAA Coordinator for this effort may be reached at:

PALS@darpa.mil

DARPA/BTO

ATTN: HR001118S0027

675 North Randolph Street Arlington, VA 22203-2114 https://www.fbo.gov/spg/ODA/DARPA/CMO/DARPA-SN-18-29/listing.html mailto:PALS@darpa.mil

PART II: FULL TEXT OF ANNOUNCEMENT

1. Funding Opportunity Description This publication constitutes an announcement from the Defense Advanced Research Projects Agency (DARPA) for a Broad Agency Announcement (BAA) process, as defined in Federal Acquisition Regulation (FAR) 6.102(d)(2) and 35.016 and 2 CFR § 200.203. Any resultant award negotiations will follow all pertinent law and regulation, and any negotiations and/or awards for procurement contracts will use procedures under FAR 15.4, Contract Pricing, as specified in the BAA. Proposals received as a result of this BAA shall be evaluated in accordance with evaluation criteria specified herein through a scientific review process.

DARPA is soliciting innovative proposals for concepts that use naturally-occurring or engineered marine organisms as sensor elements in a system designed to provide persistent monitoring of manned and unmanned vehicle (M/UUV, e.g., submarines, semi-submersibles and/or unmanned underwater gliders or vessels) movement, particularly in shallow-coastal and littoral marine environments. Specifically, DARPA seeks to identify biological organism responses to M/UUVs and develop technology to characterize and synthesize the resulting behaviors and/or signals for use as input into a detector system. The integrated technology will comprise the biological organism’s response coupled with detector hardware, software and algorithms to enable alert of the presence and movement of M/UUVs, as well as classification of size and class of the vehicle. Engineered organisms should only be used when naturally occurring organisms cannot improve upon capabilities over state of the art in terms of producing detectable signals with respect to standoff distance, feasible sensing time, and/or M/UUV classification. To the extent researchers do propose solutions that would tune organisms' reporting mechanisms, the proposers will be responsible for developing appropriate environmental safeguards to support future deployment. However, at no point in the PALS program will DARPA test modified organisms outside of contained, biosecure facilities.

1.1. PROGRAM OVERVIEW

Despite advances in sensor systems for submarine and underwater vehicle monitoring, spatial and temporal coverage of maritime environments still suffer from numerous gaps. In particular, due to a combination of high manpower requirements, lack of persistence, the need for short endurance or ship/shore-based power sources, and the often noisy or cluttered environments, current maritime procedures often employ sensor suites at the tactical level to protect high value assets, rather than at the strategic level, to maintain overall situational awareness. Furthermore, smaller targets, in particular, unmanned underwater vehicles, are exceedingly difficult to detect using current active or passive sonar technologies from any meaningful standoff distance.

Marine organisms — or natural ‘sensors’— shaped by millions of years of evolution, abound in the world’s oceans. The sensory and behavioral characteristics of these biological organisms are routinely overlooked, often treated as noise by man-made sensor users. For example, sonar operators often complain about the strong signals of snapping shrimp in reef environments, disrupting their ability to listen for targets of interest. Biological sensors occupy every maritime environment and have evolved multi-modal sensing capabilities across many domains— electrical, acoustic, optical (including in extremely low light), magnetic, chemical, and even tactile, in the form of mechanoreceptors capable of detecting and triangulating minute changes in fluid flow as well as sources of disturbances.

The Persistent Aquatic Living Sensors (PALS) program aims to leverage the biological maritime ecosystem across a wide array of marine environments, particularly in the shallow-coastal and littoral regions, to find M/UUV targets (Table 1). It aims to transform existing biology, historically characterized as background noise, into highly content-rich biological signals that can be interpreted to track, classify, and report on the presence of M/UUVs. Performers on the PALS program may consider organisms from bacteria through macro-organisms, as well as multi-organism interactions, and will:

(1) Characterize the biological signal: engineer and/or reproducibly observe, understand, and model behavioral response of biological organisms to M/UUVs and confounder objects, including discriminations of like-sized objects at multiple scales.

(2) Interpret the biological signal: detect observed unique biological signals and translate these into actionable alert information.

The PALS effort requires two stages of sensing. In the first stage, the biological organisms sense the intrusion of an M/UUV or confounder into their environment and respond with an output signal or observable behavior. In the second stage, a man-made detector system captures and interprets the unique biological signal or behavior generated by the organism(s), making an analyzed result available in the form of distilled alerts. These components will be integrated into demonstrator systems, which are able to be deployed in a maritime environment and capable of end-to-end system performance through delivery of alerts via commercial satellite link. A notional depiction of this overall vision is shown in Figure 1. Ultimately, PALS systems will offer long-endurance, widespread sensory coverage in multiple maritime environments, augmenting and enhancing current detection capabilities.

Figure 1: PALS program vision for manned/unmanned underwater vehicle detection using an amplifying biological signal transducer and associated detector, signal processor, and exfiltration platform.

1.1.1. Organisms, Targets, and Environments

Proposers are expected to enumerate the anticipated details of their system capability;

specifically, the organisms they intend to use, their target set, and the geographical region(s) associated with the overall system deployment, including both the Technical Area1 presence as well as Technical Area 2 hardware functionality. Detailed information on each technical area is provided below.

Proposers may choose from any marine organism to serve as their living sensor. Viable categories of organisms are as follows: single or clustered organisms (e.g., a fish or a school of fish), populations of organisms (e.g., a reefscape), micro-organisms, or engineered organisms (e.g., genetically modified bacteria). The organism selected must demonstrate a pronounced and characterizable response to M/UUV presence and must be called out in sufficient detail at the time of proposal to assess the likelihood of success.

PALS systems must be capable of identifying and discriminating some (or all) of the following target types or classes: 1) small Unmanned Underwater Vehicles (UUVs) with minimum dimensions (DxL) of 0.3 meters x 2.7 meters, 2) semi-submersible manned underwater vehicles with minimum dimensions (DxL) of 2.0 meters x 12 meters, or signatures (e.g., chemical, biological, optical, acoustic) associated with the construction or operation of same, 3) fully-submersible manned vehicles (aka submarines) with minimum dimensions (DxL) 2.0 meters x 30 meters up to and including nuclear-powered ballistic missile, and nuclear- or diesel-electric-powered attack submarines. Preference will be given to approaches that accurately identify and discriminate the smallest size class targets listed above.

Proposers must also identify the environment or environments where their system will be deployed, taking into consideration the prevalence of their target biological sensory organism(s), as well as the operational characteristics of these environments. Allowable operating environments are shown in Table 1 and must be at least five (5) meters in depth.

Table 1: Environment characteristics Environment Characteristics

Reef Low turbidity, shallow depths, and strong light penetration in salt water Strait Variable turbidity and light penetration in salt water River outflow/estuary High turbidity in brackish water Continental shelf/other Greater depths, lower light penetration, and large coverage areas in salt water

1.1.2. Technical Approach

The program is comprised of two technical areas (TAs). Performer teams must propose to both TAs jointly in order to produce fully integrated systems. The functionality of the integrated system will be assessed on demonstrations in maritime environments of increasing complexity.

Proposals that do not respond to both TAs will be deemed non-responsive.

Technical Area 1: Characterize Biological Signal The aim of TA1 is to study natural organisms (or possibly employ synthetic biology solutions using engineered organisms) in order to distinguish and produce detectable signals that can be observed and synthesized to produce accurate alerts of M/UUV presence. The signals are expected to generate a response that amplifies the weak signature of many M/UUV targets, whether through direct gain amplification (acoustic signature of the organism is more pronounced than the target, for example), or through modulation to another domain with a signal that propagates further or is in some way easier to detect than the original M/UUV target signal.

Synthetic biology may be pursued in order to enhance signal amplification, when necessary to achieve the proper system standoff. Experimental work on characterizing the natural biological organisms’ responses and signal outputs can start in constrained lab and tank environments and move to progressively larger, more uncontrolled maritime environments to confirm the functionality at more realistic standoff distances. The critical distance is the total standoff distance (see Figure 2), and may be apportioned between the target, biological sensor (TA1 component), and man-made detector (TA2 component) in the manner chosen by the performer – organisms can be proximal to the target with detector hardware further away, or the organisms may be far from the target with detector hardware in close proximity.

Figure 2: Standoff definitions

Engineering of new organisms will be allowed only in cases where the engineered organism offers unique sensing modalities or where the biological signal gain needs to be amplified beyond the naturally occurring variant, such that it offers sufficient content to be readily detected and analyzed, and contingent upon having built-in safeguards to prevent release into the environment. A minimum of two safeguards (e.g., encapsulation, limited nutrient supply, sensitivity to salt water, etc.) must be included as part of the envisioned concept of operations, implying that the response to target occurs with the safeguards in place. All engineered organisms must remain entirely in secure biocontainment for the duration of the program.

Engineered organisms will be tested in containment but must be designed to be deployable in the ocean without causing any deleterious effects to the environment.

In order to achieve success, performer teams will have to systematically examine their selected organisms at multiple scales (1, 10, 100, and 500 meters) and against multiple test objects (some or all of the targets defined in Section 1.1.1, with similarly sized confounders), and map very precise response functions onto target and confounder presence, reflecting a well-defined interpretation of the biological organism as a sensor transducer. This will require developing an understanding of signal roll-off as a function of organism distance to target objects, and as a function of detector distance to the organisms.

Proposers must select and justify the organism(s) they plan to use when drafting a response, along with the biological signal(s) they intend to leverage. These organisms must demonstrate the modeled behaviors in their native ecosystem. As such, proposals requiring the transportation of the selected organisms to an environment where they are not naturally found will be deemed nonresponsive, with the exception of engineered organisms. Justification should include a discussion on the worldwide prevalence of their selected organism(s), along with the transferability/heritability of the biological signals to be leveraged in different globally encountered species, genotypes, etc., essential for the functionality of the system in different maritime environments.

Technical Area 2: Interpret the Biological Signal The aim of TA2 is to detect the observed unique biological signals from TA1, and translate these into actionable information. This will require the development of the hardware, software and algorithms necessary to detect the biological organism signals; capture their unique response signals to M/UUV targets; classify the patterns associated with these responses; and use these classifications to produce distilled alerts consisting of a target’s size, type, location and bearing.

Commercial off-the-shelf host platforms must be used (see additional detail on available space, weight, and power in Appendix 2), but the detector hardware may involve novel designs in order to best capture the specific features of biological signals provided by the TA1 teams. TA2 development will occur at the same scale as TA1 research – if the organism is able to be tested in a constrained tank, the TA2 work can leverage these tanks, but larger or more distributed organisms will dictate that TA2 development be performed in more unconstrained environments.

The list of tasks to be performed by phase is presented in Table 2.

Responses to TA2 must include development of an analytical detector system able to capture and interpret the biological signal, and ensure adequate detection platform endurance. The organism defined in TA1 should provide a signal that can be captured by sensory equipment from TA2, with detail provided in Appendix 2. Performer teams must work across program phases to progressively mine more content from the biological signal, starting with object detection and ultimately working towards target classification and bearing. Software performer teams will have to increase their processing efficiency to conserve power, either through innovative code development or by leveraging industry advances over the course of the program, to support the final endurance targets.

Based on the desire to leverage the inherently ubiquitous, self-replicating nature of biological organisms as part of the overall PALS component design, TA2 hardware should not rely on manually intensive integration between the detector and the biological organism(s). Therefore, proposals involving the use of manually attached tags will be viewed less favorably than those employing proximal or remote sensing of the biological organism(s) and their behaviors. Priority will be given to systems that harness biological organisms’ ability to persistently monitor large areas of ocean more effectively than current systems, with hardware solutions that are low-profile and of suitable at-sea duration. Increasing the endurance of the detection system throughout the program is an important metric for program success.

1.1.3. Program Phases

The program will be divided into three phases, which are designed to successively build upon results of previous phases. Phases will mature the end-to-end system concept by first identifying the core biological and non-biological components, understanding system performance through quantitative testing, and progressively migrating to more challenging maritime environments in order to produce long-endurance, operationally relevant systems. At the conclusion of each program phase, testing will be conducted to confirm that each subsystem (Phase 1) or full system (Phases 2 and 3) performs as required per the specifications outlined in Section 1.2. Specifically, independent validation and verification (IV&V) testing will be conducted by a third-party organization, to be identified and retained by DARPA. This third-party IV&V organization will be responsible for establishing and executing the testing protocols and procedures in accordance with the metrics outlined in Section 1.2, with supervision from DARPA.

The three program phases will have a total duration of forty-eight (48) months. Phase 1 will be eighteen (18) months, Phase 2 will be eighteen (18) months, and Phase 3 will be twelve (12) months. At the conclusion of Phase 1, performer down-selections will occur, with continued funding being contingent upon a system’s ability to meet the required metrics and present a solid development plan for the remainder of the program, as outlined in Section 1.2. The major tasks and objectives for the program, organized by phase and TA, are provided below, in Table 2, and are discussed in more detail in the following subsections. A calendar of overall program events is outlined in Figure 3.

Figure 3: Gantt chart of major test activities by phase, with notional quarters (calendar year designation) and tasking by technical area

Table 2: Tasks by phase and technical area

Phase Technical Area 1 Technical Area 2

Phase 1 (18 months)

Characterize organism response to M/UUVs and confounder objects

Confirm target signal propagation for 1 and 10 m total standoff distances (Metrics, Table 3)

Build training datasets of organism responses

Detect biological signals in response to targets Discriminate M/UUV targets from confounders Determine expected operational system standoff

Phase 2 (18 months)

Confirm target signal propagation to biological sensor for 10, 100 and 500 m total standoff (Metrics, Table 3)

Confirm characterization of biological signal at required standoff

Test biological sensors in more realistic environment

Confirm biological signal propagation to detector for 10, 100 and 500 m total standoff (Metrics, Table 3)

Package hardware for maritime environment Refine algorithms at required standoff Develop alert reporting scheme

Phase 3 (12 months)

Optimize target characterization and discrimination by biological sensor at 100 and 500 m total standoff (Metrics, Table 3)

Adjust algorithms for real-world conditions Refine alert reporting tolerances

Phase 1 During Phase 1, performer teams will (1) evaluate the candidate organism and biological signal;

(2) characterize the organism response to the target(s); (3) develop and test the detection algorithm in a laboratory or near-shore environment; and (4) demonstrate successful observation and/or creation of the biological signal, with appropriate target discrimination (see Milestones and Deliverables by Phase, Table 4). Testing may be done in a laboratory or, for non-synthetic biology approaches, in the near-shore environment. In early Phase 1 tests, targets will be presented at a greatly reduced standoff in order to maximize the gain on the target signal. Phase 1 will contain a technical interchange meeting (TIM) across technical area groups, preliminary design reviews (PDRs) for both technical areas, and a critical design review (CDR) for the TA2 prototype.

Nine months from the formal program kickoff, performers will have a mid-phase demonstration to test their detection scheme concepts in a tank or shallow coastal environment, using benchtop laboratory instrumentation. Performers will demonstrate that their system identifies and collects the intended target signal, produces a unique biological signal in response to the targets enumerated in Section 1.1.3, and processes this signal to indicate the presence of an M/UUV.

This demonstration must prove that the prototype components offer the ability to differentiate M/UUV from confounder, including different man-made objects, environmental / organism clutter, and acoustic interference. Teams must demonstrate this ability at a total standoff of 1 meter. At month 15, performers will support an end-of-phase test and evaluation exercise managed by the IV&V team. This test must demonstrate the ability to detect 95% of all M/UUV targets with metrics outlined in Table 3. Performers will receive test results from the IV&V team by the end of month 15, and must deliver a report in month 16 containing the results and discussion of IV&V testing, as well as the proposed path to reaching metrics for Phases 2 and 3.

Performer teams who are engineering their own organisms must be able to demonstrate successful enhancement of a target signal for their organism compared to naturally-occurring variants of the organism in a controlled setting, in addition to interpreting these signals as outlined in Phase 1 and meeting the metrics as shown in Table 3. They are expected to host the mid-phase demonstration and support the IV&V end-of-phase test as described above.

Phase 2

In Phase 2, performers will continue to refine their models of organism behavior under more realistic conditions (complexity and scale), and develop and refine the hardware and software components that are required to detect and interpret the targeted biological signal(s) (see Table 4). TA1 teams will study the interactions of their organisms with both targets and confounders to develop models of behavioral differences, performed in larger and more complex environments to produce more complete datasets. Meanwhile, TA2 performers will develop their detection and analysis system to fit onto a commercial off-the-shelf (COTS) platform, such as a buoy. The space, weight, and power available for system components are presented in Appendix 2, which includes considerations for system components designed for operation from the seafloor. Phase 2 will include a technical interchange meeting (TIM) across technical areas, as well as preliminary and critical design reviews (PDR / CDR) for the TA2 system. Performers will undertake large seawater tank or aquarium-style simulated ecosystem or marine tests using their integrated system to improve detection and confidence levels and to demonstrate performance in a larger, more realistic environment. The simulated ecosystem must resemble the target environment as closely as possible to enable performers to adequately discriminate the target from naturally occurring confounders. Performers must use experimentally derived data to produce a receiver operating characteristic (ROC) curve on overall system performance, illustrating the performance trade-off between false alarm and target detection. Performers will provide their ROC curves as a living document to be refined during Phase 3, including open-ocean testing, and must present a risk table that identifies the confounder organisms and/or phenomena that are most likely to interfere with identification of the M/UUV target in open ocean tests, along with a mitigation strategy for each. They will also host a mid-phase demonstration to confirm their progress.

The overall system will need to report on target size and bearing, and must deliver alerts within ten minutes of detection by the TA2 hardware. A packaged hardware system must be provided to the IV&V team for a 30-day maritime deployment test during which the hardware must operate without intervention (detail in Section 1.2). In addition, performers are expected to support an end-of-phase IV&V test event, which will ensure that all metrics (Table 3) are met.

Performer teams who are engineering their own organisms must be able to demonstrate systems (e.g., discriminate target from confounders, larger standoff distances) consistent with Phase 2 metrics, milestones, and deliverables in a biocontainment facility. Teams working with engineered organisms must also demonstrate their two safeguard systems by the end of Phase 2, in advance of large-scale testing in Phase 3.

Phase 3 In the final phase, teams must migrate their system to an uncontrolled marine environment, and increase their total standoff and system endurance significantly. The Phase 3 development environment must naturally contain the organisms or ecosystems that the teams intend to use as their biological sensing elements, with the exception of synthetic biology approaches. Teams must demonstrate that their systems satisfy the metrics shown in Table 3, including testing out to 500 meters total standoff, while handling more confounding targets as would be found in real-world deployment (see Milestones and Deliverables by Phase, Table 4). TA1 teams will focus on enhancing their behavioral models and testing the organism responses at the increased standoff, while TA2 teams will focus on receipt of biological signal, increased computational efficiency, and more accurate reporting detail. Performer teams who are engineering their own organisms will continue to work in constrained environments using their two environmental safeguards, but will also need to demonstrate increased sensitivity and specificity at 100 m in the presence of confounders. The final integrated hardware and software must maintain a 60-day endurance in a marine environment at a duty cycle of one measurement every thirty minutes. At the conclusion of Phase 3, performers will support end-of-phase testing managed by the IV&V team in a marine environment, will confirm that all metrics (Table 3) are met.

1.2. PROGRAM METRICS

In order for the Government to evaluate the effectiveness of a proposed solution in achieving the stated program objectives, proposers should note that the Government hereby promulgates the following program metrics that may serve as the basis for determining whether satisfactory progress is being made to warrant continued funding of the program. Continued funding for each subsequent phase is contingent upon meeting or exceeding the metrics prescribed for the current phase. Although the following program metrics are specified, proposers should note that the Government has identified these goals with the intention of bounding the scope of effort, while affording the maximum flexibility, creativity, and innovation in proposing solutions to the stated problem, to include variations in performance across the target class size.

The metrics identified for this program were designed with the intent to encourage and drive creative and innovative solutions that lead to incremental increases in the understanding of the biological responses being exploited, and the utility of these signals in the detection and discrimination of potential targets over the course of the program. Further, these metrics serve to increase the suitability and effectiveness of the overall detection system. The metrics identified, for each TA, apply to the system in its entirety, and will require that both TA-specific objectives are met in order for the metrics to also be successfully met. Only successful completion of both TAs will be considered successful completion of a given phase. Overarching program metrics are outlined in Table 3.

Table 3: Program metrics

Metric Phase 1 Phase 2 Phase 3

Object Detection at System Standoff1 95% at 1 meter 95% at 10 meters 95% at 100 meters 1 meter 0.60 N/A – Not Tested N/A-Not Tested

10 meters 0.25 0.90 N/A – Not Tested 100 meters N/A – Not Tested 0.70 0.90

Sensitivity

(TP/P)

500 meters2 N/A – Not Tested 0.25 0.70 1 meter 0.60 N/A – Not Tested N/A-Not Tested

10 meters 0.25 0.90 N/A – Not Tested 100 meters N/A – Not Tested 0.70 0.90

Specificity

(TN/N)

500 meters2 N/A – Not Tested 0.25 0.70 Hardware endurance N/A 30 Days 60 Days Report timeliness3 N/A 10 Minutes 5 Minutes

Report content Target size Target bearing +/- 22.5 Target bearing +/- 15

1. System standoff is the sum of target-to-organism distance and organism-to-detector distance, defined in Figure 2. 2. Teams using genetically modified organisms must follow all metrics in a constrained environment and will only test out to 100 m. 3.

“Timeliness” is defined as the amount of time elapsed between detection of a potential target (i.e. by the man-made detector) and the system’s discrimination between a target and a confounder and the issuance of an alert, when appropriate.

The object detection metric will require that 95% of all presented objects, both M/UUV and similarly size confounders, trigger a response from the biological sensor and a subsequent detection from the TA2 system, across all three size classes presented in Section 1.1.1, at the distances presented in Table 3. When the presentation of actual targets is not feasible (for example, use of a full-sized manned submarine), the IV&V team will utilize a surrogate target able to produce a similar signal intensity at the specified distance, with this surrogate target mutually agreed upon between DARPA, the IV&V team, and the performer as based on organism and detector phenomenology employed.

Sensitivity and specificity metrics refer to the overall system’s ability to accurately capture the correct signals and reject erroneous signals (false alarms, including both improperly classified confounder objects as well as alerts when no objects are present). Both parameters will be calculated by the IV&V team as part of their end-of-phase test analysis. The detailed calculation of these metrics is presented in Appendix 3. The metrics presented are for the smallest targets to be identified by the proposed system, with the expectation that standoff will vary with target size. DARPA expects performers to propose detail on the distance where their sensitivity and specificity may fall below (or exceed) the metrics provided for different sized targets, and will consider that in the evaluation process.

For hardware endurance metrics, one full copy of the TA2 detector and alerting system will be surrendered to the Government IV&V team, to be deployed into a controlled seawater environment for the durations specified in Table 3. The system will not be presented with actual M/UUV targets during this testing window, but rather will be pre-configured by the TA2 team to be presented with a simulated biological sensor alert consisting of pre-recorded data that is known to engage the entire processing pipeline of the TA2 detector and alerting systems, once every 2 hours. The system will be continuously sensing as if it were operationally deployed, and should send alerts of simulated target detection throughout the test in order to demonstrate continued functionality. Upon conclusion of the test window, the system must be fully functional, as demonstrated by the IV&V team.

The reporting timeliness metric is intended to ensure near real-time processing of the biological signals by the detector system, as opposed to extensive post-processing. Furthermore, some organisms may demonstrate a response with cumulative effects, which reach a threshold level for emitting a detectable signal that has a slow time constant. Therefore, the calculation of the timeliness metric is defined as the time from receipt of the biological signal at the detector system to the time that an alert is sent via satellite link to the system user. The IV&V team will coordinate with performers to determine the best manner in which to timestamp the detector system receipt time.

Alert reporting content must contain basic information in all phases of the program – timestamp, detector GPS location, and target object size (from the 3 bins of sizes outlined in Section 1.1.1).

In addition, in Phases 2 and 3, the system must include a detected object’s bearing from the detector system, with precision requirements called out in Table 3.

Proposals should cite the quantitative and qualitative success criteria that the effort will achieve by the time of each phase’s program metric measurement.

1.3. PROGRAM MILESTONES AND DELIVERABLES

In addition to the specific metrics outlined in sub-section 1.2 above, the following milestones must be provided. These milestones are specific to each TA and phase, and reaching these milestones will facilitate overall, successful completion of the requirements for this program.

These milestones, along with the corresponding deliverables, are identified in Table 4.

Table 4: Milestones and deliverables by phase

Phase Technical Area Milestones Deliverables

Tech Area 1

Technical interchange meeting 1 Preliminary design review of organism selection / engineering Characterize response of organism(s) to target or target proxy Produce training data sets for TA2 algorithm development Host mid-phase demonstration Facilitate IV&V test to quantify detection performance and target discrimination at 1 and 10 m standoff

For engineered organisms only:

demonstrate superior performance over naturally occurring organisms

For engineered organisms only: Month 2:

Phase 1 organism design

Month 7: Mid-Phase Demonstration test plan

Month 10: Mid-Phase Demonstration report (30 days after demo)

Month 12: Report containing risk table and mitigation plan

Month 16: Report on Phase 1 IV&V test results and Phase 2 task plan (30 days after test)

Phase 1 (18

Months)

Tech Area 2

Preliminary design review of hardware/software prototypes

Critical design review of hardware/software prototypes

Demonstrate benchtop hardware and software developed to perform detection

Construct initial algorithm to discriminate target signal from noise

Generate design for packaged system to be constructed in Phase 2

Month 6: Final prototype design Month 15: One (1) prototype system of biological organisms and necessary hardware / software to detect biological signals provided to IV&V team

Month 16: Code report on algorithms used to characterize biological signal

Tech Area 1

Technical interchange meeting 2 Characterize organismal response to target in presence of confounders Produce training data sets in simulated ecosystem environment Host mid-phase demonstration Facilitate IV&V test for sensitivity and specificity at 10, 100 and 500 m For engineered organisms only:

demonstrate 2 safeguard systems

Month 20: Environmental assessment and permit documents

Month 24: Mid-phase demonstration plan Month 28: Mid-phase demonstration report Month 32: Report containing the ROC curve and risk table with mitigation plan Month 35: Report on Phase 2 IV&V test results and Phase 3 task plan (30 days after demo)Phase 2

(18 Months)

Tech Area 2

Preliminary design review of hardware/software system

Critical design review of hardware/software system

Construct refined algorithms based on simulated ecosystem data

Package hardware system for 30-day marine endurance test

Develop target bearing algorithm Produce alert reporting scheme

Month 24: Phase 2 system design Month 34: Provide two (2) fully integrated systems to IV&V team for endurance and performance testing

Month 35: Updated code report on algorithms used to characterize biological signal

Phase Technical Area Milestones Deliverables

Tech Area 1

Technical interchange meeting 3 Characterize organismal response to target in real world environment Facilitate IV&V tests for sensitivity and specificity at 100 and 500 m

Month 38: Updated environmental assessment and permit documents

Month 47: Final report on IV&V test results (30 days after test)

Phase 3 (12

Months) Tech

Area 2

Produce or adapt hardware for 60-day endurance test

Develop refined algorithms based on real world environment data

Develop improved bearing scheme Produce improved alert reporting scheme

Month 45: Provide two (2) fully integrated systems to IV&V team for endurance and performance testing

1.4. PROGRAM DEMONSTRATIONS

In order to ensure that appropriate progress is being made towards meeting the prescribed metrics, teams will be required to demonstrate their systems—or, when applicable, specific sub-systems—to the Government, at pre-defined times during and/or just prior to the conclusion of a given program phase. Each demonstration or test will have a specific purpose related to the particular goals of the given phase, as described below.

Phase 1 Midterm Demonstration At the midpoint of Phase 1 (i.e. no later than month 9), a benchtop/laboratory-scale demonstration will be hosted by the performer. Teams will be required to present the initial results of the efforts to characterize the biological signal of interest and must demonstrate the ability to identify M/UUV targets and discriminate target from confounder at the artificially-close standoff of one meter. By the time of the midterm demonstration, teams will be required to prove that a qualitatively unique biological signal is generated in response to a potential target, and that the signal generated in response to a man-made target is distinct from those generated by confounders (e.g., floating debris, marine life, etc.). These differences can occur in the signal content, timing, or any other domain suitable to recognize a distinction between an M/UUV and a confounder.

End-of-Phase IV&V Test During month 15, each PALS approach will be tested, and systems must fully satisfy the Phase 1 metrics identified in Table 3. Testing will be performed by a third-party, independent verification and validation (IV&V) team, to be retained by the Government, and will be supported by the performer teams as-needed. The IV&V team will confirm that systems are capable of detecting and discriminating targets at multiple standoff distances, as identified in Table 3. Teams must surrender their hardware, software, and (if necessary) may hand over organisms to the IV&V team, to have them perform the testing.

Phase 2 Midterm Prototype Demonstration At the midpoint of Phase 2 (i.e. no later than nine (9) months from Phase 2 start date), individual system prototypes will be demonstrated by the performer in a surrogate environment (large aquarium, salt water tank or coastal area), and must demonstrate improved sensitivity, specificity, and range from Phase 1 performance. The demonstration must present a clear path towards achieving the final Phase 2 targets. This demonstration will be hosted by the performer.

End-of-Phase IV&V Testing and Demonstration Near the end of Phase 2, a test will be managed by the IV&V team and supported by the performer as-needed. For this event, two systems must be surrendered to the IV&V team to test performance against the Phase 2 metrics of Table 3 – one system for endurance testing, and one for performance testing. Performance testing will be performed in a surrogate environment (aquarium, large salt water tank or coastal area). The system must also demonstrate that it can perform its intended functions during a 30-day endurance test in a marine environment. The DARPA team will coordinate with the IV&V team to determine an appropriate demonstration window during this testing, and may invite Government stakeholders to observe the event at their discretion. Teams with engineered organisms will be tested in constrained environments that simulate real-world conditions and must demonstrate both safeguards to the IV&V team in this constrained test environment.

Phase 3 At the conclusion of Phase 3, fully integrated systems will be tested by the IV&V team in a real-world, marine environment, and must be capable of detecting and discriminating targets and delivering operationally relevant alerts. For this event, two systems must be surrendered to the IV&V team to test performance against the Phase 3 metrics of Table 3 – one system for endurance testing, and one for performance testing. Performer teams are expected to support the IV&V-led testing, as-needed. Engineered organisms will be tested in a controlled environment for total standoff distances as large as 100 meters, with environment to be identified by the IV&V team in close coordination with DARPA and the performer team. The system must also demonstrate that it can withstand a 60-day endurance test in a marine environment. DARPA will coordinate with the IV&V team to determine an appropriate demonstration window during system testing, and may invite Government stakeholders to observe the event at their discretion.

There is no mid-phase demonstration in Phase 3.

1.5. GENERAL REQUIREMENTS

Proposing Teams It is expected that proposals will involve multidisciplinary teams that include expertise from multiple complementary disciplines (e.g., biology, chemistry, engineering, weak signals detection, machine learning, data science, analytics, oceanography, and physics).

Specific content, communications, networking, and team formation are the sole responsibility of the proposer teams. Proposer teams must submit a single, integrated proposal led by a single

Program Integrator/Manager or prime contractor that addresses all program phases, as applicable.

DARPA will hold a Proposers Day (see Section 8, Other Information) to facilitate the formation of proposer teams with the expertise necessary to meet the goals of the program and enable sharing of information among interested proposers through the DARPA Opportunities Page and the Proposers Day registration website.

Data Sharing DARPA anticipates that a large amount of data will be generated under this program by each performer and that the analyses and validation will be strengthened by compiling and integrating information across all performers. The PALS program will require that performer data, analysis, and software executables (or source code) be shared with DARPA, the IV&V team, and US Government stakeholders. Performers are strongly encouraged to establish the appropriate agreements to enable collaboration and data sharing beyond these organizations. DARPA encourages sharing of pre-existing data, including those generated through funding from other sources, although this is not a requirement of the program.

Biocontainment/Bio Safety (Engineered Organisms) This program will support engineered biological research conducted in containment and will not support proposals that include uncontained environmental release of engineered organisms.

Controlled Unclassified Information (CUI) To prevent the release of sensitive technical information, certain aspects of proposals may be considered CUI and may require safeguarding or dissemination controls, pursuant to and consistent with applicable laws, regulations, and Government-wide policies to include Department of Defense Manual 5200.01 Volume 4. CUI as defined is not classified under Executive Order 13526 or the Atomic Energy Act, as amended.

Common categories of CUI at DARPA include the following:

Controlled Technical Information (CTI) - Distribution B through X;

Export Controlled and International Traffic in Arms Regulations (ITAR);

International Agreements;

Privacy - Personally Identifiable Information (PII);

Procurement & Acquisition - Proprietary Information;

Procurement & Acquisition - Source Selection Sensitive;

Law Enforcement Sensitive (LES); and DoD Category - For Official Use Only (FOUO).

Certain military technical information relating to the sensors and testing environment in real-world scenarios and relationships between both biological sensors and non-biological sensors and their targets could also be considered CUI by DARPA. Final determinations will be made on a case-by-case basis.

Ship Time Request (if applicable)

University-National Oceanographic Laboratory System (UNOLS)-based ship time will be paid for by DARPA, and should not be factored into performer budget estimates. However, to ensure adequate resourcing for this and to aid in evaluation, proposers must present ship time needs in their responses. These needs will not include ship time associated with the end-of-phase IV&V testing, as this ship time will be managed by the IV&V team instead of the performers. Based on the target environments listed in Table 1, Coastal/Local Class vessels should be sufficient for most project activities. Larger, as well as non-UNOLS, vessels may be considered at the discretion of the Government team. Ship time information is requested in the abstract submission as described in the instructions in section 4.2.1 (below – Proposal Abstract Format), and in the full proposals in the form of submitting a preliminary UNOLS ship time request (STR) to reserve the vessel(s) contingent on selection. To provide the UNOLS STR, select DARPA as the agency in the pulldown menu, add all other appropriate information, and manage the requests per the UNOLS process found at: https://strs.unols.org/Public/diu_login.aspx. Non-UNOLS ship time should be factored into performer budgets.

Permits and compliance It is the proposing team’s responsibility to obtain all necessary federal, state, and local government permits and approvals, and abide by all applicable laws where necessary for the proposed work to be conducted. If work is to be conducted in international waters, all international laws must be followed. Proposing teams are expected to design their proposals so that they minimize the potential adverse impact on the environment. Proposals will be reviewed to ensure that they have sufficient environmental documentation to allow the Government team to determine whether the proposal is categorically excluded from further National Environmental Policy Act (NEPA) analysis, or whether an Environmental Assessment or Environmental Impact Statement is necessary in conformance with requirements of the NEPA. For those applications needing an Environmental Assessment, affected proposers will be informed after the proposal review stage and will be requested to assist in the preparation of a draft of the assessment (prior to award). Failure to apply for and/or obtain federal, state, and local permits, approvals, letters of agreement, or failure to provide environmental analysis where necessary (e.g., NEPA environmental assessment) will also delay the award of funds if a project is otherwise selected for funding.

Other Requirements Performers are expected to attend program level meetings to provide scientific and technical updates to the selected performers on the PALS program on progress towards their milestones and scientific goals, and to summarize outstanding challenges and limitations that must still be overcome to achieve the overarching goals of the program. Program level meetings will be held at the kick-off of each phase (Phases 1, 2 and 3), as well as mid-phase (Phases 1 and 2) for the program duration.

https://strs.unols.org/Public/diu_login.aspx

2. Award Information

2.1. GENERAL AWARD INFORMATION

Multiple awards are possible. The amount of resources made available under this BAA will depend on the quality of the proposals received and the availability of funds.

The Government reserves the right to select for negotiation all, some, one, or none of the proposals received in response to this solicitation and to make awards without discussions with proposers. The Government also reserves the right to conduct discussions if it is later determined to be necessary. If warranted, portions of resulting awards may be segregated into pre-priced options. Additionally, DARPA reserves the right to accept proposals in their entirety or to select only portions of proposals for award. In the event that DARPA desires to award only portions of a proposal, negotiations may be opened with that proposer. The Government reserves the right to fund proposals in phases with options for continued work, as applicable.

The Government reserves the right to request any additional, necessary documentation once it makes the award instrument determination.