Detect and Avoid Systems: Navigating the Present and Shaping the Future

The recent Detect and Avoid (DAA) workshop, alongside panel discussions and a standards overview presented at the UAS Integration events in May 2025, offered a rich and nuanced perspective on the progress made by industry and regulators in safely integrating unmanned aircraft systems (UAS) into diverse and complex airspace environments. These discussions highlighted both significant advancements and critical gaps that still require attention.

A central theme highlighted is the industry's significant advancement in standardizing test methodologies through organizations such as ASTM F38 and RTCA SC-228. Frameworks like the "wagon wheel" encounter geometry provide robust, replicable test setups that enable operators to gather extensive empirical data. These frameworks serve as guidelines, offering a structured approach to validate and benchmark system performance against various intruder scenarios. At Martin Solutions, LLC, we've effectively utilized these standards not as a strict rulebook, but as flexible guidelines that inform our design and validation approaches, firmly grounded in the operational realities we face.

Yet, while the standards have provided vital technical scaffolding, operational experience demonstrates that they are not comprehensive solutions by themselves. A particularly stark example of this is the absence of an explicit FAA definition for the critical term "well clear." Although widely referenced in regulatory contexts, "well clear" remains deliberately undefined by the FAA due to its inherently context-dependent nature. This regulatory ambiguity poses a significant validation challenge. However, at Martin Solutions, we've leveraged this ambiguity as an opportunity: defining our safety buffers based on rigorous encounter simulations and tailored detection and avoidance strategies specifically designed to meet or exceed the intent of maintaining safe separation.

Another critical realization is that automation in Detect and Avoid is not monolithic; rather, it exists along a spectrum tailored to specific operational contexts. Some operations rely on "pilot-in-the-loop," where systems detect and alert, but pilots retain full decision-making responsibility. Others transition towards "pilot-on-the-loop," where automated avoidance maneuvers trigger if pilots fail to respond in time—ideal for dense airspace environments where pilot saturation is a concern. Still, others advance toward full autonomy—"pilot-out-of-the-loop"—a stage essential for scalability that brings unique certification and trust challenges. Clearly articulating the appropriate level of automation and integrating it with realistic human factors considerations, pilot interfaces, and response timing is vital for achieving operational safety and regulatory acceptance.

Closely linked to automation strategies is the emerging role of Electronic Conspicuity (EC). EC, including ADS-B and Remote ID, enables aircraft to broadcast their position and intent, significantly enhancing strategic mitigation capabilities. EC shows particular promise in densely populated urban areas, low-altitude Class G airspace, and environments with a high percentage of non-cooperative aircraft. However, despite its potential, the integration of EC into DAA strategies remains inconsistent, partly due to varying regulatory requirements and differences in fleet equipage. At Martin Solutions, LLC, we have observed the practical benefits of strategically combining cooperative (EC-based) and non-cooperative detection technologies to improve overall situational awareness and mitigation capabilities. We believe EC should transition from an optional element to a foundational pillar within detect-and-avoid architectures.

However, the discussions highlighted several persistent gaps: a universal operational matrix aligning automation and EC capabilities with airspace and mission types; rigorous human factors testing customized to automation levels; and comprehensive guidance on integrating terrain-aware avoidance maneuvers into automated logic. Equally important is the absence of a strong framework for post-event logging, data analytics, and system forensics, which are essential for continuously improving and validating DAA performance.

In practice, the success of any DAA system depends on a clear understanding of operational objectives, identifying key operational variables, and explicitly managing these variables. At Martin Solutions, we have found that real-world validation, driven by mission-specific encounter modeling, performance-based safety arguments, and iterative testing, consistently outperforms rigid adherence to standards alone.

As the industry advances, it must adopt operational diversity, simplify human-system interfaces, strategically integrate EC, and firmly ground DAA implementations within mission-specific constraints. The DAA state-of-the-art is promising, yet a significant collaborative effort remains necessary. By recognizing and thoughtfully addressing these gaps, especially the FAA's intentional lack of a universal "well clear" definition, the industry can drive toward a safer, scalable, and truly integrated future for unmanned aviation.

Ultimately, Detect and Avoid represents more than just a system: it's a constantly evolving operational philosophy, shaped by context, enhanced by technology, and guided by the clarity of our safety objectives.