- Details
- Category: Patent Analysis
The U.S. Department of the Air Force has licensed a patented drone-capture technology to Rocket One Inc. under a 12-month evaluation agreement announced on October 6, 2026.
The agreement covers U.S. Patent No. 10,239,637, titled “System and Method for Arresting and Neutralizing Unmanned Vehicles”. Granted in March 2019, the patent describes a system designed to physically capture unmanned vehicles while isolating their wireless communications.
Rocket One intends to evaluate the technology for potential defense and security applications, including possible integration with its Swarm Stage AI counter-unmanned aircraft systems (counter-UAS) initiatives.
The Backstory: Why the Air Force Developed the Technology
The invention emerged during a period of growing concern over the security risks posed by inexpensive commercial drones.
In January 2015, a small quadcopter crashed on the White House grounds, exposing challenges associated with protecting sensitive facilities from unauthorized unmanned aircraft.
Later that year, the Department of Defense conducted its Black Dart 2015 demonstration at Naval Base Ventura County, California. The exercise evaluated counter-drone technologies and highlighted the difficulty of detecting small UAVs using conventional radar systems.
Against this broader security backdrop, Air Force inventors Jonathan Ashdown, Paul Sikora, and Brendon Poland developed an approach addressing a limitation of existing drone-capture technologies: physically trapping a drone did not necessarily prevent its operator from transmitting commands or activating a hazardous payload.
Their invention, which claims priority to March 3, 2016, proposed combining physical containment with radio-frequency isolation to reduce these risks without necessarily destroying the target.
Although these events illustrate the environment in which the invention emerged, there is no established evidence that the White House incident or Black Dart exercise directly inspired the inventors.
How the Patented Drone-Capture System Works
The patent describes a friendly unmanned aerial vehicle (UAV) deploying a specialized containment structure around a potentially hostile drone.
The enclosure incorporates conductive shielding material intended to attenuate radio-frequency signals, functioning similarly to a Faraday cage. By limiting signals entering or leaving the enclosure, the system could disrupt communications between the captured drone and its operator.
Patent drawing FIG. 1: Cross-section of the patented RF isolation material, showing conductive shielding and supporting layers.
The conductive shielding works on principles associated with a Faraday enclosure, attenuating electromagnetic signals that would otherwise reach equipment inside. In a drone-interception scenario, this could reduce the effectiveness of external radio-control commands and wireless positioning signals.
The patent describes two principal containment configurations: a deployable net-like enclosure and a substantially rigid cage. In either configuration, the friendly drone approaches the target and deploys the containment structure around it.
Patent drawing FIG. 3
An especially significant aspect involves electronic countermeasures. As illustrated in Figure 2 of the ’637 patent, the system incorporates an external antenna (201) and an internal antenna (203) positioned on opposite sides of the RF-shielding material. These antennas could support signal interception, controlled communications, or localized spoofing and jamming within the shielded environment. (see more on antenna insights)
Patent drawing FIG. 2: Cross-sectional view of the RF-isolating containment structure, showing internal and external antennas and their associated signal cables.
Related Patents: How Counter-Drone Capture Technology Has Evolved
The Air Force invention forms part of a broader body of patent activity involving drone interception, electromagnetic protection, and electronic countermeasures.
Three other patents illustrate how this technology has developed:
 |
Granted in 2015 and assigned to Lockheed Martin, this patent describes an aerial platform equipped with a deployable net for capturing another UAV. It is specifically discussed in the Air Force patent's background as earlier capture technology that did not address RF isolation in the same way. |
 |
U.S. Patent No. 11,054,224 - System for Physically Capturing and Signal-Defeating Unmanned Aerial Vehicles This later patent describes a capture system incorporating a conductive signal-blocking enclosure and an additional capture device. It is especially relevant because it also combines physical interception with RF shielding, although its particular claimed arrangement differs from the Air Force patent. |
 |
U.S. Patent Application Publication No. 2024/0142996 - Counter Drone System This published application addresses networked detection, sensor-data fusion, identification of potential drone threats, and instructions to counter-drone systems. It illustrates a complementary layer of technology that could inform decisions about when and how an interception system should be deployed. |
These patents illustrate the evolution of counter-drone systems from mechanical interception toward integrated physical and electronic neutralization. They are technologically related, although they do not necessarily share ownership or licensing rights.
What the License Could Mean for AI-Enabled Counter-Drone Systems
Rocket One plans to evaluate whether the patented technology could complement its Swarm Stage AI platform, which focuses on drone-swarm threat simulation and counter-UAS training.
Potential applications could combine AI-assisted detection, target tracking, and interception planning with physical capture and RF isolation. However, the licensed patent itself does not primarily claim AI technology, and these integrations remain prospective rather than demonstrated.
The agreement grants evaluation rights only, not commercial sales authorization. Further testing, licensing, and regulatory approvals would be required before commercialization.
Conclusion
U.S. Patent No. 10,239,637 illustrates an approach to counter-drone technology that brings together mechanical interception, electromagnetic shielding, and electronic countermeasures.
Its licensing to Rocket One highlights how government-developed intellectual property may contribute to emerging AI-enabled defense applications, particularly as unmanned systems become increasingly sophisticated.
About Kama Thuo, PLLC
This analysis is provided by Kama Thuo, PLLC, an engineering & technology law firm focused on patents, AI, and wireless telecom law. Brian Kibet is a multidisciplinary professional with expertise in wireless engineering, paralegal practice, and software development. With a strong background in AI and automation, he designs intelligent workflows for intellectual property processes, with a focus on patent-related work, bridging the gap between technical innovation and legal strategy.
Whether you are an inventor seeking to license your technology or a company navigating an IP dispute, our firm has the technical and legal expertise to protect your interests. Reach out to us to see how we can assist you at www.kthlaw.com/patents, or explore our AI-powered legal services at https://www.kthlaw.com/ai.
- Details
- Category: Patent Analysis
Development platforms for building custom AI solutions in patent law and legal technology. These tools enable patent attorneys and IP professionals to create, deploy, and maintain AI-powered applications for patent analysis, document automation, and legal workflow optimization.
LangChain LangChain
A comprehensive framework for building applications with large language models. For patent AI tools, LangChain excels at creating document processing pipelines, integrating multiple data sources (patent databases, legal documents), and chaining together different AI operations like patent classification, prior art search, and claim analysis. Its extensive library of pre-built components makes it ideal for rapid prototyping of patent workflow automation.
LangGraph LangChain
A more recent addition to the LangChain ecosystem that focuses on building stateful, multi-agent workflows with explicit control flow. In patent law applications, LangGraph is particularly valuable for complex multi-step processes like patent prosecution workflows, where you need different AI agents handling distinct tasks (claim drafting, prior art analysis, office action responses) with conditional logic and state management between steps. Whereas LangChain is better for linear, pipeline-based patent workflows, while LangGraph excels at complex, branching processes requiring state management and agent coordination.
CrewAI CrewAI
An open-source Python framework for creating and managing collaborative AI agent systems. CrewAI can orchestrate patent teams where different AI agents can handle distinct roles, for example, a prior art specialist, a claim drafting expert, a prosecution strategy advisor, and a client communication manager. This role-based architecture mirrors traditional patent law firm structures, enabling sophisticated workflows like coordinated patent family management across multiple jurisdictions, where AI agents collaborate to ensure claim consistency, deadline compliance, and strategic prosecution approaches. The framework's independence from other platforms provides maximum control over sensitive patent processes.
Skyvern Skyvern
An open-source browser agent that can enable patent professionals, and other professionals, to automate work on websites. Using LLMs and computer vision, Skyvern can log in to filing portals, fill in forms, and download documents, even when a site has no API. It also works with n8n, so a web step can join an existing workflow.
Vertex AI Agent Builder Google
Google Cloud's enterprise-grade platform for building scalable AI agents with robust security and compliance features essential for legal applications. The Agent Garden provides pre-built templates which can be used for common patent tasks, while the RAG capabilities enable AI agents to ground responses in authoritative patent databases and patet law legal precedents. Particularly valuable for large patent portfolios, Vertex AI can create AI agents that perform real-time patent landscape analysis, automated prior art searches across global databases, and client support automation for patent status inquiries. Its enterprise security features ensure compliance with attorney-client privilege and data protection requirements.
n8n n8n
n8n provides tools for building AI agents through its workflow automation platform. This low-code workflow automation platform can enable patent professionals to build AI agents integrated with existing patent management systems and legal software. The AI Agent node can be configured to automate routine patent tasks like docket entry processing, client status updates, and preliminary patent searches. n8n can enable connecting patent AI capabilities with established legal technology stacks, enabling workflows that automatically process patent applications, extract key information for docket systems, and trigger appropriate follow-up actions. This integration capability makes it ideal for firms wanting to enhance existing patent processes with AI without replacing current systems.
LlamaIndex LlamaIndex
A framework focused on data connection and retrieval for AI agents. LlamaIndex agents are well suited for context-augmented patent analysis, retrieving relevant information from vast patent corpora to perform comprehensive prior art searches, technical claim analysis, and patent landscape mapping. Its RAG capabilities enable agents to access real-time patent data from multiple sources (USPTO, EPO, WIPO) while maintaining context about specific client technologies and competitive landscapes. Workflows can combine multiple patent analysis AI agents to create sophisticated research and reporting systems for complex patent matters.
Strands Agents Amazon
An open-source SDK that can enable patent law firms to build sophisticated AI agents with minimal code complexity. By leveraging state-of-the-art language models' native reasoning and tool-use capabilities, Strands allows legal professionals to create AI agents that can automate patent research, document analysis, prior art searches, and case management workflows. The model-driven approach means patent attorneys can focus on defining their specific legal requirements rather than wrestling with complex orchestration logic.
Lindy Lindy
A no-code AI agent platform that can enable patent professionals to build customized automation without technical expertise. Lindy can create patent workflow agents that can manage patent docket deadlines, automate client communications regarding prosecution status, and coordinate multi-step patent filing processes. Its knowledge base integration allows agents to be trained on firm-specific patent procedures, USPTO guidelines, and client preferences. Multi-agent capabilities can enable complex patent prosecution workflows where different agents handle prior art searches, claim analysis, and deadline management while sharing information seamlessly.
Zep zep
A specialized memory layer that can transform patent AI agents from reactive tools into intelligent assistants with institutional knowledge. For patent law applications, Zep can enable AI agents to build comprehensive knowledge graphs of client patent portfolios, prosecution histories, and examiner behavior patterns over time. This temporal knowledge capability is crucial for patent strategy, allowing AI agents to track how patent claims evolve through prosecution, identify successful argument patterns with specific examiners, and maintain context across multi-year patent families. Zep's integration with frameworks like LangChain allows patent law professionals to build AI agents that learn and improve from each interaction.
- Details
- Category: Patent Analysis
Specialized AI-powered tools designed for patent drafting, patent analysis, claim charting, prior art searches, patent prosecution, portfolio management, and licensing workflows. These tools streamline key workflows and enhance accuracy, efficiency, and strategic insight in patent practice.
IPRally IPRally Technologies Inc
AI patent search and classifier uses knowledge graphs and Graph Neural Networks. Patent classifier allows users to use their own taxonomy by retraining the custom classifier and applying it to any data set or linking it to a monitoring profile for automated technology insights. For patentability, can define invention in plain text or use their graph editor.
PQAI PQAI
A collaborative, not-for-profit initiative to build an open-source ecosystem of AI components to drive innovation and improve patent quality.
Amplified Amplified AI, Inc
Searching in Amplfied Projects works by combining AI similarity sorting with Boolean filtering. Amplified understands both conceptual and semantic similarity. Mark results as relevant to find more conceptually similar patents. Turn learning off to prefer semantic similarity to your text only.
Patseer PatSeer Technologies Pvt. Ltd.
IP research platform for AI-enabled patent search, due diligence, litigation tasks, and business development. Integrate patents and legal status data feeds. Offers integrated analytics, project workflow, and collaboration capabilities.
Powerpatent Powerpatent, Inc
Patent prosecution tool to streamline invention disclosure capture, flowchart & drawing management, graphical claim drafting, computer-aided description drafting, diagnostics to catch §112 (enablement) and claim issues, and inventor/client collaboration.
Ambercite Ambercite
Paid tool focuses on prior art searching, licensing, patent landscaping, portfolio benchmarking.
NLPatent NLPatent
Copy-and-paste an invention disclosure, abstract or set of claims directly into the app, or describe the invention in your own words; similarity model ranks results in order of conceptual relevancy; specify mandatory criteria that your results list must exhibit, such as certain keywords, ownership, legal status, priority date; tool constantly learning from your input through intelligent refinement; export results into a spreadsheet or a professionally formatted report. Watch Video.
Octimine Dennemeyer
AI patent analysis platform. Combines boolean and AI-powered semantic search. Includes automated monitoring system for continuous updates on changes to legal status and additions to technology field.
Xlscout XLSCOUT
Xlscourt is an integrated innovation and patent monetization platform. Xlscout's ClaimChart LLM utilizes Large Language Models to identify potential licensees and their overlapping products and provides AI-generated claim charts and Evidence of Use (EoU). Watch Video.
- Details
- Category: Patent Analysis
Priority Date
The priority date is perhaps the most critical date in patent law because it sets the benchmark against which the novelty and inventiveness of the invention are evaluated. The priority date gives the applicant priority over others for the same invention, meaning that subsequent disclosures or filings by others concerning the same or a substantially similar invention will not affect the novelty of the originally filed invention. The priority date could be:
- The date on which an inventor first files a patent application for an invention, either in a national patent office or under a treaty that the country recognizes, such as the Paris Convention or the Patent Cooperation Treaty (PCT).
- The filing date of an earlier application in which the current application "claims priority to."
Filing Date
The filing date is the date on which a patent office receives a patent application that meets the basic filing requirements. For national applications, this is simply the date the application is submitted to the patent office. In the context of a PCT application, the initial filing date under the PCT serves as the international filing date and can also serve as the priority date if it is the first filing for that invention.
371(c) Date
The 371(c) date, also known as the national stage entry date, specifically pertains to international patent applications filed under the Patent Cooperation Treaty (PCT). The PCT process allows an inventor to file one international patent application to seek protection for an invention in multiple countries simultaneously. Each PCT application goes through an international phase, during which it is assessed by the International Searching Authority.
After the international phase, to proceed in each designated or elected state, the application must enter the "national phase" in each of those jurisdictions. The "371(c) date" refers to the date on which the international application transitions into this national phase in a particular jurisdiction. According to PCT Article 22 or 39(1), applicants must fulfill certain requirements by this date, which might include payment of national fees and submission of translations of the application. This date is crucial as it marks the continuation of the patent prosecution process in each of the national patent offices where protection is sought.
Issue Date (or Grant Date)
This is the date on which the patent is officially granted by the patent office. From this date, the patent owner obtains the exclusive rights to exclude others from making, using, selling, or importing the patented invention.
Publication Date
Most patent applications are published 18 months after their earliest filing or priority date (unless a non-publication request is made). The publication date is when the details of the application become publicly accessible, allowing the public to begin assessing the potential scope and impact of the patent once granted.
Maintenance Fee Due Dates
Patent owners must pay maintenance fees (also known as renewal fees or annuities) at regular intervals to keep the patent in force. These fees are typically due several years after the issue date and continue throughout the life of the patent. Failure to pay maintenance fees results in the patent expiring (although it can be revived depending on how much time has elapsed since the maintenance fee was due).
Expiration Date
This date marks the end of the patent's term, after which the patented invention enters the public domain. In most countries, the standard patent term is 20 years from the filing date of the application, assuming all maintenance fees are paid. However, this can be extended under certain circumstances, such as regulatory review periods for pharmaceutical patents. Following the AIA, patent terms can be adjusted (via Patent Term Adjustment (PTA)) due to delays in the patent office in examining the patent.
PCT National Phase Entry Deadlines
For PCT applications, typically, the deadline to enter the national phase in designated countries is 30 months from the earliest priority date. This period can vary slightly depending on the regulations of the particular national patent office.
About the Author
Roque Thuo (KF7RCQ) is a licensed attorney (Arizona, California, and USPTO) and a licensed professional electrical engineer (Arizona) specializing in wireless telecommunications law. He advises clients on engineering and technology matters, including patent drafting, analysis, portfolio management, AI technology counseling, and wireless telecommunications law. He combines legal and hands-on technical expertise to help clients navigate complex IP & technology transaction challenges. Prior to founding his law firm, Roque practiced patent law at an Am Law 50 firm, following earlier hands-on experience as an integrated circuit design engineer at Fortune 500 high-tech companies.
Contact Roque today for expert legal and engineering guidance.
- Details
- Category: Patent Analysis
USPTO kind codes are two-letter codes used by the United States Patent and Trademark Office (USPTO) to identify the kind of patent document and its publication status. These codes are appended to the patent number and help to distinguish between different types of patent documents such as utility patents, design patents, plant patents, and others, as well as their status, like whether they are published applications or granted patents.
Here's a breakdown of some common USPTO kind codes used for utility patents:
|
WIPO ST.16 Kind Codes
|
Kind of document
|
Comments
|
|
A1
|
Patent Application Publication
|
Pre-grant publication available March 2001
|
|
A2
|
Patent Application Publication (Republication)
|
Pre-grant publication available March 2001
|
|
A9
|
Patent Application Publication (Corrected Publication)
|
Pre-grant publication available March 2001
|
|
B1
|
Issued Patent
|
No previously published pre-grant publication
|
|
B2
|
Issued Patent
|
Having a previously published pre-grant publication and available March 2001
|
|
C1, C2, C3
|
Reexamination Certificate
|
Previously used codes B1 and B2 are now used for granted Patents
|
|
E
|
Reissue Patent
|
No change
|
These kind codes are part of a broader international system used by many patent offices, known as the "kind code system," which facilitates the identification and categorization of patent documents globally.
See USPTO for additional information.
About the Author
Roque Thuo (KF7RCQ) is a licensed attorney (Arizona, California, and USPTO) and a licensed professional electrical engineer (Arizona) specializing in wireless telecommunications law. He advises clients on engineering and technology matters, including patent drafting, analysis, portfolio management, AI technology counseling, and wireless telecommunications law. He combines legal and hands-on technical expertise to help clients navigate complex IP & technology transaction challenges. Prior to founding his law firm, Roque practiced patent law at an Am Law 50 firm, following earlier hands-on experience as an integrated circuit design engineer at Fortune 500 high-tech companies.