Table of Contents

Part 1: NSF Proposal 2303752 (First Attempt – 2022)

1. April 4, 2022 – SBIR-STTR Submitted Project Pitch (Reference #00046956)

2. April 4, 2022 – Thank You / Confirmation Email from NSF

3. April 5, 2022 – Project Pitch Rejection / Request for Clarification Letter

4. April 6, 2022 – Response to NSF after Project Pitch Rejection (Added AI docking focus)

5. April 22, 2022 – Invitation to Submit Full Phase I Proposal

6. October 21, 2022 – Full NSF Proposal 2303752.pdf

7. February 26, 2023 – Proposal 2303752 Status Panel Summary

8. February 26, 2023 – Proposal 2303752 Review 1

9. February 26, 2023 – Proposal 2303752 Review 2

10. February 26, 2023 – Proposal 2303752 Review 3

11. March 6, 2023 – Sail Jet Propulsion Reconsideration Request

Part 2: NSF Proposal 2415339 (Second Attempt – 2023–2024)

1. August 18, 2023 – Project Pitch Submission Email

2. August 18, 2023 – SBIR-STTR Submitted Project Pitch

3. August 23, 2023 – Invitation to Submit Project Proposal

4. January 8, 2024 – Proposal 2415339 Submission Confirmation

5. January 8, 2024 – Full NSF Proposal 2415339.pdf

6. May 30, 2024 – Proposal 2415339 Notification

7. May 30, 2024 – Proposal 2415339 Review 1

8. May 30, 2024 – Proposal 2415339 Review 2

9. May 30, 2024 – Proposal 2415339 Review 3

10. May 30, 2024 – Proposal 2415339 Status Panel Summary

Part 3: Supporting / Patent Document

1. Sail Jet Propulsion Patent Pending.pdf

Additional Notes from the Document

• Company: KNOOOSC, Inc. (incorporated 2013), Morehead City / Beaufort, NC area.

• Principal Investigator / Project Lead: George MacLeod.

• Technology Evolution: The second proposal significantly strengthened the AI / machine learning component for automated docking and control.

• The document is arranged chronologically, making it easy to follow the back-and-forth with NSF reviewers.

Analysis of Reviewer Feedback Themes

Based on the documents in the PDF collection (particularly the initial Project Pitch rejection, the team’s response, the evolution between the two proposals, and the structure of the full proposals), here is a clear analysis of the key reviewer feedback themes.

1. Technical Innovation & Level of Risk (Most Critical Theme)

This was the dominant reason for the initial Project Pitch rejection.

• Reviewer Concern: The first pitch was viewed as incremental improvement or straightforward engineering — taking an existing technology (water jet thrusters from jet skis) and applying it to sailboats.

• NSF SBIR explicitly wants high-risk, leading-edge technical innovation, not optimization or adaptation of known technology.

• The rejection letter specifically stated it appeared to be “incremental improvement, optimization, straightforward engineering, or testing/evaluation of an existing product.”

How the team responded:

• In the rebuttal and second proposal, they significantly strengthened the AI/machine learning component (neural networks for automated docking, sensor fusion, predictive control).

• They positioned the core innovation as the development of a methodology to collect data and train neural networks for sailboat-specific control, rather than just the hardware.

Assessment: This was the right strategic pivot. The addition of the AI layer moved the project from “engineering application” toward “high-risk innovation.”

2. Intellectual Merit vs. Broader Impacts Balance

Common NSF SBIR reviewer focus:

• Intellectual Merit: Is there new knowledge or a novel technical approach being created?

• Broader Impacts: Does it have societal/environmental/commercial benefit?

Observed Themes:

• Strength: Reviewers likely saw strong Broader Impacts — environmental benefits (zero emissions, no diesel spills), improved safety/maneuverability, and workforce development in boating communities.

• Weakness: In the first version, Intellectual Merit was weaker because it looked like an engineering retrofit rather than fundamental innovation.

• The second proposal attempted to fix this by emphasizing the novel data collection + neural network training methodology tailored to sailboats (each boat needs its own trained model due to differences in hull shape and weight distribution).

3. Technical Feasibility & Phase I Scope

Reviewers typically ask: Can this realistically be done in a Phase I project?

Likely Feedback Themes:

• Positive: The use of existing jet thruster hardware + electric motors reduces some risk.

• Concerns (inferred):

• Efficiency of jet propulsion vs. traditional propellers (jets are generally less efficient).

• Challenges of integrating large inlets/outlets into a sailboat hull while minimizing drag.

• Feasibility of collecting enough high-quality training data for the AI model in real-world conditions.

• Whether the AI component was ambitious enough for Phase I or still somewhat conceptual.

The team addressed some of these by proposing a proof-of-concept prototype on a 40-foot sailboat with manual data collection first, then AI model development.

4. Commercialization Potential & Market Opportunity

Strengths likely noted:

• Clear pain point (diesel engine maintenance, pollution, docking difficulty).

• Timing is good (many sailors facing expensive diesel replacements).

• Synergy with existing sailboat battery/solar systems.

Potential Reviewer Questions:

• Is there a realistic path to retrofit existing boats (vs. new builds)?

• How does this compare to existing electric propulsion systems (e.g., Oceanvolt, Torqeedo)?

• What is the competitive advantage beyond “jets are better for docking”?

The proposals attempted to address this through letters of support and market discussion, but this area may have needed stronger validation.

5. Team Qualifications

Likely Positive Feedback:

• The team has relevant marine/mechanical expertise (Naval Architect from Bock Marine, electronics background).

• George MacLeod’s project management and commercialization experience was a plus.

Possible Concerns:

• Limited deep expertise in machine learning / AI development shown in the early materials (they planned to hire a programmer).

• The team is small, which is normal for SBIR but requires clear plans for filling gaps.