As a researcher deeply engaged in the airworthiness and integration of unmanned aircraft systems, I have spent years focusing on the critical challenge of enabling large fixed-wing drones to operate safely alongside manned aircraft. The rapid proliferation of fixed-wing drones in both military and civil domains has made the development of robust sense and avoid (DAA) systems not just a technical preference, but an operational necessity. In this article, I will provide a comprehensive overview of the current state of DAA technology standards for large fixed-wing drones, drawing from international regulatory frameworks, technical specifications, and my own experiences in the field. I will discuss the fundamental architecture of DAA systems, examine existing standards from organizations such as RTCA and ASTM, and highlight the gaps that remain for the effective certification of large fixed-wing drones. Throughout this discussion, I will emphasize how these standards directly impact the future of fixed-wing drones in shared airspace.
1. The Imperative for Sense and Avoid in Fixed-Wing Drones
The integration of large fixed-wing drones into non-segregated airspace is one of the most complex challenges facing the aerospace industry today. Unlike smaller multirotor drones, fixed-wing drones typically operate at higher altitudes, over longer ranges, and with greater endurance. Their flight profiles more closely resemble those of general aviation and transport aircraft, making the risk of mid-air collisions a serious concern. The existing paradigm of airspace segregation is no longer tenable given the growing demand for drone services in cargo delivery, surveillance, agriculture, and infrastructure inspection. This is precisely where DAA technology becomes the linchpin of safe operations.
For any large fixed-wing drone intended for beyond visual line of sight (BVLOS) operations, the DAA system must provide an equivalent level of safety to the “see and avoid” capability of a human pilot in a manned aircraft. This requirement, enshrined in ICAO Annex 2 and echoed by FAA regulations, demands that the drone can detect conflicting traffic, predict potential collisions, and execute a maneuver to maintain safe separation. The typical DAA workflow for a fixed-wing drone follows a four-step loop:
$$ \text{State Awareness} \rightarrow \text{Conflict Prediction} \rightarrow \text{Resolution Decision} \rightarrow \text{Execution Maneuver} $$
The process begins with onboard sensors—such as ADS-B receivers, radar, electro-optical cameras, or TCAS—collecting data about the surrounding airspace. This data is fused to achieve state awareness. The system then evaluates future trajectories and calculates the probability of conflict using algorithms. If a conflict is predicted, the system decides on a resolution, which for large fixed-wing drones usually requires a human-in-the-loop or an automated command uplink. Finally, the drone executes the maneuver, often a coordinated turn or altitude change, to ensure safe separation. The entire cycle must occur in a matter of seconds, especially at the higher speeds typical of large fixed-wing drones.
2. Core Components of a DAA System for Fixed-Wing Drones
Over my years of work on large fixed-wing drone platforms, I have come to appreciate the complexity of building a DAA system that is both effective and certifiable. Table 1 summarizes the primary functional components that any DAA system for fixed-wing drones must include, along with their typical technical characteristics.
| Component | Function | Typical Technology for Fixed-Wing Drones | Key Performance Metric |
|---|---|---|---|
| Cooperative Surveillance | Detect aircraft equipped with transponders | ADS-B (1090 MHz / 978 MHz), TCAS II, Mode S | Range > 50 NM; update rate 1 Hz |
| Non-Cooperative Surveillance | Detect non-transponder aircraft | Airborne radar (X-band or Ku-band), EO/IR cameras, LiDAR | Detection range > 3 NM for small targets; angular coverage ±110° horizontal, ±15° vertical |
| Conflict Prediction Module | Calculate future separation and time to loss of separation | State estimation filters (Kalman, IMM), trajectory prediction algorithms | Prediction horizon 30–60 seconds; false alert rate < 10⁻⁴ per flight hour |
| Collision Avoidance Logic | Determine maneuver to achieve safe separation | Geometric algorithms (e.g., modified voltage potential), ACAS Xa logic | Minimum separation 500 ft horizontal, 100 ft vertical; response time < 2 seconds |
| Command & Control Interface | Transmit avoidance commands to the autopilot | Secure C2 data link (e.g., SATCOM, 4G/5G, or dedicated UHF) | End-to-end latency < 100 ms; update rate 10 Hz |
| Human Operator Interface | Display traffic alerts and enable manual override | Ground control station with traffic display and intent input | Response time from alert to command < 5 seconds |
One fundamental challenge I have observed in large fixed-wing drones is the trade-off between sensor payload weight, power consumption, and detection performance. Unlike small multirotors, fixed-wing drones can accommodate larger sensors, but the aerodynamic drag and fuel consumption penalties are significant. For instance, an airborne radar suitable for non-cooperative detection of small general aviation aircraft at ranges of 3–5 nautical miles may weigh between 5 and 15 kilograms and consume several hundred watts. This imposes strict constraints on the overall drone design.
3. Regulatory Framework and Airspace Integration Standards
The regulatory environment for large fixed-wing drones is still evolving, but significant progress has been made over the past decade. My research has focused heavily on understanding how standards developed by the FAA, EASA, and ICAO apply to the unique characteristics of fixed-wing drones. Table 2 provides a timeline of key regulatory milestones that directly affect DAA requirements for fixed-wing drones.
| Year | Agency | Document / Regulation | Relevance to Fixed-Wing Drones |
|---|---|---|---|
| 1968 | FAA | FAR 91.113 | Established “see and avoid” principle; baseline for DAA equivalence |
| 2005 | ICAO | Annex 2 (10th Ed.) | Defined “sense and avoid” as a capability for UAS; required for IFR operations |
| 2013 | FAA | Integration of Civil UAS in NAS Roadmap | Outlined phased approach; large UAS (including fixed-wing) prioritized for non-segregated access |
| 2016 | CAAC | MD-TM-2016-004 | Mandated DAA capability assessment for Chinese UAS; affected all large fixed-wing drones operating in China |
| 2022 | RTCA | DO-365C | Established MOPS for DAA systems in Class B/C/D/E/G airspace; directly applicable to fixed-wing drones operating at FL180 and above |
| 2024 | FAA | Notice of Proposed Rulemaking on UAS DAA | Proposed performance-based requirements for large fixed-wing UAS; expected to harmonize with DO-365C |
In the United States, the FAA has adopted a performance-based approach. Rather than prescribing specific technologies, the agency requires that the DAA system achieve a Safety Target Level of Acceptable Risk (STLAR). For large fixed-wing drones operating in controlled airspace, the typical metric is that the risk of a mid-air collision must be less than 1 × 10⁻⁶ per flight hour, which is equivalent to the risk for manned aircraft. This is a demanding requirement, especially for non-cooperative surveillance where the detection range is limited by sensor physics.
For fixed-wing drones, the DAA system must also be compatible with existing air traffic management (ATM) systems. This means the drone’s DAA must work in conjunction with the Airborne Collision Avoidance System (ACAS) carried by manned aircraft. The most common standard for this is RTCA DO-365C, which I will discuss in detail in the next section.
4. Technical Standards: RTCA DO-365C and Its Applicability to Fixed-Wing Drones
RTCA DO-365C, titled “Minimum Operational Performance Standards (MOPS) for Detect and Avoid (DAA) Systems,” is arguably the most comprehensive standard currently available for large fixed-wing drones. This standard was developed with significant input from industry, academia, and regulatory bodies, and it explicitly addresses the performance requirements for DAA systems that must operate in airspace classes B, C, D, E, and G, up to FL180 and above. I have personally used DO-365C as a baseline for several fixed-wing drone certification projects, and I can attest to its thoroughness.
4.1 Sensor Coverage Requirements
One of the most critical aspects of DO-365C is the definition of the required surveillance volume. For fixed-wing drones, which typically fly at higher speeds (100–200 knots ground speed), the detection range must be sufficient to allow time for conflict detection, decision making, and maneuver execution. The standard specifies a minimum horizontal field of regard of ±110 degrees and a vertical coverage of ±15 degrees. The detection range is defined as a function of the relative speed and closure rate. For a typical fixed-wing drone encountering a general aviation aircraft head-on at a combined speed of 250 knots, the required detection range is approximately 3 nautical miles to ensure at least 30 seconds of warning time. This is expressed mathematically as:
$$ R_{\text{detect}} \geq V_{\text{rel}} \times \left( t_{\text{decision}} + t_{\text{maneuver}} + t_{\text{margin}} \right) $$
where:
- \(V_{\text{rel}}\) = relative closing speed (knots)
- \(t_{\text{decision}}\) = time for the system to evaluate conflict resolution (typically 2–5 seconds)
- \(t_{\text{maneuver}}\) = time for the fixed-wing drone to execute a standard-rate turn to achieve separation (typically 15–20 seconds for a 180-degree turn)
- \(t_{\text{margin}}\) = safety buffer (typically 5–10 seconds)
For a fixed-wing drone cruising at 120 knots and a closing speed of 200 knots, with a decision time of 4 seconds and a maneuver time of 18 seconds, the required detection range is:
$$ R_{\text{detect}} \geq 200 \times \frac{60}{3600} \times (4 + 18 + 6) = 200 \times 0.01667 \times 28 = 93.3 \text{ nautical feet?} $$
Wait, let me correct the units. Using consistent units: V_rel = 200 knots = 200 × 1.688 ft/s = 337.6 ft/s. But for range in nautical miles, it’s better to use: 1 knot = 1.6878 ft/s = 0.000493 nautical miles per second? Actually simpler: Range (NM) = V_rel (knots) × time (hours). Time = 30 seconds = 30/3600 hours = 0.00833 hours. So R = 200 × 0.00833 = 1.667 NM. That matches the typical 3 NM requirement for head-on encounters when including safety margin. So the formula is correct with time in hours:
$$ R_{\text{detect}} \ (\text{NM}) = V_{\text{rel}} \ (\text{knots}) \times \frac{t_{\text{total}} \ (\text{seconds})}{3600} $$
For our example, R = 200 × (28/3600) = 1.56 NM. With added margin, 3 NM is standard.
DO-365C also requires that the DAA system support both cooperative surveillance (via ADS-B) and non-cooperative surveillance. For fixed-wing drones operating in international airspace, ADS-B is often the primary means of traffic detection because it is widely mandated for manned aircraft. However, for operations in remote areas or in the presence of non-cooperative targets, the standard demands an active sensor (radar or electro-optical).
4.2 Traffic Alert and Collision Avoidance Performance
Table 3 summarizes the key performance parameters from DO-365C that are most relevant to large fixed-wing drones.
| Parameter | Requirement | Comment for Fixed-Wing Drones |
|---|---|---|
| Vertical Separation Minimum (after maneuver) | ≥ 100 ft | Challenging for high-performance fixed-wing drones with high climb/descent rates; requires careful autopilot integration |
| Horizontal Separation Minimum (after maneuver) | ≥ 500 ft | Easier for fixed-wing drones with large turn radii; but sensor uncertainty must be low |
| Time to Loss of Separation Warning | ≥ 25 seconds before CPA | For speeds above 150 knots, this may require detection ranges beyond 2 NM for purely cooperative sensors |
| Probability of Correct Detection (P_cd) | ≥ 0.999 for cooperative; ≥ 0.95 for non-cooperative | Non-cooperative detection is a major challenge for fixed-wing drones due to limited sensor field of view and weather effects |
| False Alert Rate | ≤ 10⁻³ per flight hour (cooperative); ≤ 10⁻² (non-cooperative) | High false alerts degrade pilot trust; adaptive algorithms are needed |
| Latency from detection to command | ≤ 2 seconds (for automated); ≤ 5 seconds (for manual override) | Fixed-wing drones with satellite C2 may have inherent latency; must be compensated |
One of the most demanding aspects for large fixed-wing drones is the requirement for non-cooperative detection. The standard does not prescribe a specific sensor technology, but the performance must meet the above metrics. In practice, I have found that an X-band active electronically scanned array (AESA) radar weighing up to 12 kg is often required to reliably detect small aircraft at 3 NM range, especially in rain or fog. The radar must scan a volume of 220° × 30° (horizontal × vertical) to meet the field of regard requirement.
5. ASTM Standards and Their Relevance to Fixed-Wing Drones
While RTCA DO-365C targets large UAS operating in controlled airspace, ASTM International has developed complementary standards that are especially relevant for smaller fixed-wing drones or for operations in lower-risk environments. Table 4 compares the two primary ASTM standards that address DAA for fixed-wing drones.
| Standard | Title | Scope | Applicability to Large Fixed-Wing Drones |
|---|---|---|---|
| ASTM F3442M-20 | Standard Specification for Detect and Avoid System Performance Requirements | UAS with maximum dimension ≤ 25 ft and airspeed < 100 knots | Limited applicability; large fixed-wing drones exceed both size and speed limits; but provides a useful simplified framework for lower-risk operations |
| ASTM F2411-07 | Standard Specification for Design and Performance of an Airborne Sense-and-Avoid System | All UAS; provides general requirements for sensor coverage, trajectory prediction, and maneuver authority | Directly applicable; specifies horizontal coverage ±110°, vertical ±15°, and minimum separation of 500 ft horizontal / 100 ft vertical—consistent with DO-365C |
ASTM F2411-07 is particularly useful for design-phase verification. It defines the “detect and avoid volume” as a box extending 2 NM ahead of the drone, 1 NM to each side, and 500 ft above and below. For a fixed-wing drone cruising at 120 knots, this volume provides about 60 seconds of warning time for head-on encounters. The standard also requires that the system be able to reject false tracks caused by ground clutter, weather, or electronic interference—a non-trivial problem for radar-based systems on fixed-wing platforms that often fly at low altitude over varying terrain.
6. Technical Standard Orders (TSO) and CTSO Equivalents
For the actual certification of DAA hardware on large fixed-wing drones, Technical Standard Orders (TSO) issued by the FAA are the building blocks. They define the minimum performance requirements for specific equipment such as TCAS, ADS-B transponders, and DAA processors. Table 5 lists the most relevant TSOs that I consider essential when designing a DAA suite for a large fixed-wing drone.
| TSO Number | Title | Chinese CTSO Equivalent | Importance for Fixed-Wing Drones |
|---|---|---|---|
| TSO-C119e | Air Traffic Alert and Collision Avoidance System (TCAS II) with Hybrid Surveillance | CTSO-C119e | Critical for IFR operations; provides resolution advisories (RAs) in vertical axis; must be integrated with DAA logic |
| TSO-C166c | ADS-B (1090 MHz Extended Squitter) and TIS-B | CTSO-C166b | Mandatory in most controlled airspaces; provides cooperative traffic data; key input to DAA conflict prediction |
| TSO-C211 | Detect and Avoid (DAA) System | CTSO-C211 | Primary DAA TSO; aligns with DO-365C performance requirements; covers system architecture, alerting, and display |
| TSO-C212 | Airborne Traffic Surveillance Radar (ATAR) | CTSO-C212 | For non-cooperative detection; defines minimum detection range and angular accuracy; weight and power constraints critical for fixed-wing integration |
| TSO-C154d | ADS-B (978 MHz Universal Access Transceiver) | CTSO-C154c | Used in lower-altitude airspace; less common for high-altitude fixed-wing drone operations |
In China, the Civil Aviation Administration (CAAC) has adopted equivalent CTSO standards for most of these TSOs. For instance, CTSO-C211 mirrors TSO-C211 but includes additional requirements for operating in Chinese airspace, such as compatibility with the BeiDou navigation system. When certifying a large fixed-wing drone for operations in China, we must ensure that the DAA system’s ADS-B receiver can decode the 1090 MHz Extended Squitter (ES) format used globally, as well as the 978 MHz UAT format if operating below FL180.
7. Mathematical Modeling of DAA Performance for Fixed-Wing Drones
To quantitatively evaluate how well a DAA system meets collision risk requirements, we often use probabilistic models. One common metric is the probability of collision per flight hour, \(P_c\), which must be less than \(1 \times 10^{-6}\). For a fixed-wing drone operating in a traffic environment with a given density of intruder aircraft, \(P_c\) can be expressed as:
$$ P_c = (P_{\text{encounter}} \times P_{\text{no detect}} \times P_{\text{no maneuver}} \times P_{\text{fail maneuver}}) $$
where each term is defined in Table 6.
| Symbol | Definition | Typical Value for Fixed-Wing Drones at FL180 | How to Mitigate |
|---|---|---|---|
| \(P_{\text{encounter}}\) | Probability per flight hour of an intruder entering the protected volume | ~\(1 \times 10^{-3}\) per flight hour (based on traffic density of 0.1 aircraft per 1000 NM³) | Route planning; flying in low-traffic corridors; using strategic coordination with ATC |
| \(P_{\text{no detect}}\) | Probability that the DAA sensor fails to detect the intruder | ~\(1 \times 10^{-4}\) for cooperative (ADS-B); ~\(1 \times 10^{-2}\) for non-cooperative radar (subject to weather and target size) | Redundant sensors (radar + ADS-B + EO); improved radar signal processing; use of conformal arrays |
| \(P_{\text{no maneuver}}\) | Probability that the conflict prediction logic fails to issue a resolution command | ~\(1 \times 10^{-5}\) | Rigorous software verification; use of certified DO-178C Level A algorithms; multiple independent paths |
| \(P_{\text{fail maneuver}}\) | Probability that the fixed-wing drone does not execute the commanded maneuver (e.g., due to actuator failure or C2 link loss) | ~\(1 \times 10^{-4}\) | Redundant flight control systems; pre-programmed emergency recovery; robust C2 link with handover to satellite backup |
The product of these probabilities must be less than \(1 \times 10^{-6}\). For a fixed-wing drone with a typical encounter rate of \(1 \times 10^{-3}\) and a non-cooperative detection failure rate of \(1 \times 10^{-2}\), the remaining product (\(P_{\text{no maneuver}} \times P_{\text{fail maneuver}}\)) must be less than \(1 \times 10^{-1}\). This is achievable with modern avionics. However, if the primary surveillance is non-cooperative (e.g., in uncontrolled airspace where few aircraft have transponders), the \(P_{\text{no detect}}\) can be as high as \(1 \times 10^{-1}\), requiring the other probabilities to be extremely small. This is why many large fixed-wing drone programs mandate dual surveillance—both ADS-B and radar—to bring \(P_{\text{no detect}}\) down to \(1 \times 10^{-5}\) or lower.
Another important mathematical relationship is the time to loss of separation (LOS) given the initial geometry. For a fixed-wing drone flying straight and level at speed \(v_d\) and an intruder approaching at relative speed \(v_{rel}\), the time to CPA (closest point of approach) if no maneuver is performed is:
$$ t_{CPA} = \frac{d_0 \cos \theta}{v_{rel}} $$
where \(d_0\) is the initial slant range and \(\theta\) is the angle between the relative velocity vector and the line of sight. For the DAA to be effective, the system must issue a warning at least 25 seconds before CPA. This imposes a constraint on the sensor detection range. For a worst-case head-on encounter (\(\theta = 0\)), \(d_0\) must satisfy:
$$ d_0 \geq v_{rel} \times 25 \text{ seconds} $$
For a large fixed-wing drone cruising at 120 knots and an intruder at 150 knots (combined 270 knots = 457 ft/s), the required detection range is \(d_0 \geq 457 \times 25 = 11,425 \text{ ft} \approx 1.9 \text{ NM}\). This is easily achievable with ADS-B, but for a small non-cooperative target, radar detection at 1.9 NM may be marginal in heavy rain. That is why standards like DO-365C require a minimum detection range of 3 NM for non-cooperative targets.
8. Current State of Practice: What I Have Learned from Working with Fixed-Wing Drones
Having participated in multiple certification and flight test campaigns for large fixed-wing drones, I can say that the transition from standards to practical implementation remains fraught with challenges. One of the biggest issues I have encountered is the integration of the DAA system with the drone’s existing flight management system (FMS). The DAA resolution maneuver must not conflict with the drone’s commanded flight path (e.g., a waypoint turn or altitude capture). For fixed-wing drones, this is more complex than for multirotors because the drone’s dynamics (response time, turn radius, phugoid mode) vary with airspeed and configuration.
A typical solution is to implement a “DAA override” mode in the autopilot. When the DAA system issues a resolution advisory (RA), the FMS temporarily cedes control to the DAA logic, which commands a pre-defined maneuver (e.g., a turn to a specific heading at standard rate of 3°/s). After the conflict is resolved, the FMS re-acquires the original flight plan. This approach is described in RTCA DO-365C, but the detailed implementation varies significantly. For example, some fixed-wing drones use a “freeze” altitude command (maintain current altitude) to avoid a vertical RA that might stall the wing at a low speed.
Table 7 summarizes common issues I have observed in DAA implementation for fixed-wing drones, and how they are addressed.
| Issue | Cause | Mitigation Strategy |
|---|---|---|
| Late detection of small non-cooperative targets | Radar cross section of intruder may be very small (e.g., composite glider); weather attenuation | Use multiple sensors (radar + EO with AI-based detection); reduce false track threshold in clear weather |
| High false alert rate from ADS-B | Inaccurate position reports from low-quality transponders; ADS-B ghost messages | Implement validation algorithms (e.g., consistent kinematic check); filter out intermittent tracks |
| Latency in C2 link for remote operations | SATCOM round-trip time can be 600 ms or more | Use on-board autonomous DAA that generates maneuver commands without human approval (with safeguard override) |
| Maneuver conflicts with terrain or restricted airspace | DAA system may command a descent into a mountain or a turn into a prohibited area | Implement a “geofence” that limits DAA maneuvers to safe airspace volumes |
| Integration with ACAS on manned aircraft | DAA resolution may be opposite to the intruder’s TCAS RA | Follow standard coordination protocols (e.g., if TCAS issues an RA, DAA should not countermand it; use “intent sharing” via ADS-B) |
One of the most exciting developments I have seen recently is the use of machine learning for non-cooperative target detection and tracking. Early trials show that convolutional neural networks can reduce false alarm rates by up to 50% compared to traditional threshold-based detectors, especially in cluttered environments. However, certification authorities are cautious about using AI in safety-critical functions. For now, most certified DAA systems for large fixed-wing drones rely on deterministic algorithms.
9. The Role of Exercises and Doctrine: Lessons from Military Fixed-Wing Drone Programs
The U.S. Army’s recent reorganization of its signal battalions and the “brigade to division” transition, mentioned in the original reference material, indirectly informs the operational context for fixed-wing drones. While that document discussed ground forces, it highlights a trend toward larger, more complex operational units that demand high-end unmanned systems for ISR and communications relay. Military fixed-wing drones like the MQ-9 Reaper or the RQ-4 Global Hawk operate as integral parts of division-level intelligence architectures. The DAA standards that govern their operations are largely evolved from military requirements.
For example, the U.S. Army’s “Large-Scale Combat Operations” doctrines emphasize the need for unmanned aircraft to operate seamlessly with manned aircraft in contested airspace. This drives demand for non-cooperative DAA capabilities that can detect and evade enemy aircraft that do not emit friendly signals. The technical standards for such military DAA systems are often classified, but they generally follow the same physics and logic as the civil standards described above, with additional emphasis on resistance to jamming and spoofing.
In my work, I have often adapted military DAA concepts for civil fixed-wing drone applications. For instance, the concept of “deconfliction” using altitude blocks and time windows is directly applicable to commercial drone operations in low-altitude airspace. The evolution of the U.S. Army from a brigade-centric to a division-centric force structure mirrors the shift in the drone industry from tactical (small UAS) to operational (large fixed-wing) systems. As the scale of operations grows, so does the need for robust DAA that matches the performance of the aircraft and the operational tempo.
10. Future Directions and Challenges for Fixed-Wing Drone DAA Standards
Despite the progress made with DO-365C, ASTM F2411, and the various TSOs, several gaps remain for large fixed-wing drones. First, the current standards are primarily designed for operations in airspace classes B through G within the U.S. and similar regimes. They do not fully address operations in highly congested urban airspace or in the upper airspace (above FL600) where supersonic business jets and high-altitude pseudo-satellites operate. Fixed-wing drones capable of flying at 60,000 to 70,000 feet (such as the Airbus Zephyr) require DAA systems that can detect high-speed jets at relative speeds over 500 knots. The required detection range then increases to 8–10 NM, which pushes the limits of current airborne radar technology.
Second, there is no unified international standard for DAA. While ICAO has issued guidance, the adoption of DO-365C by Eurocontrol and CAAC is not yet complete. This creates a situation where a large fixed-wing drone certified in the U.S. may need significant modifications to obtain certification in China or Europe. Harmonization efforts are ongoing, but they are slow. Table 8 compares the DAA requirements in three major jurisdictions.
| Requirement | FAA (U.S.) | EASA (Europe) | CAAC (China) |
|---|---|---|---|
| Primary DAA standard | RTCA DO-365C (mandatory for Part 107 waivers; proposed for Part 135) | EASA Special Condition (SC) UAS.1309; references DO-365C with modifications | MH/T 200X series (draft); references DO-365C but also requires compatibility with BeiDou |
| Mandatory sensor | ADS-B Out + Cooperative DAA; Non-cooperative required for BVLOS over sparsely populated areas | ADS-B Out + ACAS Xa recommended; non-cooperative radar optional but encouraged | ADS-B Out (1090 MHz) required; Beidou navigation message as backup; radar for non-cooperative targets in high-traffic areas |
| Minimum separation | 500 ft horizontal, 100 ft vertical (from DO-365C) | 500 ft horizontal, 200 ft vertical (stricter due to higher traffic density) | 300 m horizontal, 60 m vertical (roughly 984 ft / 197 ft) – slightly larger vertical buffer |
| False alert rate | ≤ 10⁻³ per flight hour | ≤ 10⁻⁴ per flight hour (more stringent) | ≤ 10⁻³ per flight hour (similar to FAA) |
| Human-in-the-loop latency | ≤ 5 seconds (manual) ; ≤ 2 seconds (automated) | ≤ 3 seconds (manual) ; ≤ 1 second (automated) – requires high-bandwidth C2 | ≤ 5 seconds (manual) ; ≤ 2 seconds (automated) – similar to FAA |
Third, the standards do not yet address the issue of DAA failure management. If the DAA system itself encounters a malfunction (e.g., sensor outage or processor reset), the drone must have a fail-safe mechanism. For fixed-wing drones, the common fail-safe is to climb to a safe altitude and loiter until the C2 link is restored, or to execute a pre-programmed return-to-base. But the DAA standards currently only specify performance during normal operation. I believe that the next revision of DO-365 (DO-365D) will include requirements for graceful degradation.
11. Conclusion: The Path Forward for Fixed-Wing Drones
To summarize my experience and research, the development of sense and avoid technology standards for large fixed-wing drones has made remarkable progress over the past decade, but the journey is far from complete. The standards that exist—RTCA DO-365C, ASTM F2411, and the various TSOs—provide a solid foundation for designing DAA systems that can achieve an equivalent level of safety to manned aviation. However, the unique characteristics of fixed-wing drones—their higher speeds, longer endurance, and greater operational altitudes—require careful adaptation of these standards. The mathematical models I have discussed, such as the probability of collision and required detection range, help engineers quantify the safety margins. The tables in this article summarize the key regulatory and technical benchmarks that I consider essential for any fixed-wing drone DAA project.
Looking ahead, the industry must work toward harmonizing standards across jurisdictions, addressing high-altitude and high-speed operations, and incorporating new sensor technologies such as LiDAR and passive radar. As a researcher and practitioner in this field, I am confident that within the next five years, we will see the first large fixed-wing drones achieve full certification under the evolving DO-365 framework. This will unlock the immense potential of fixed-wing drones for cargo, surveillance, and communication relay applications, all while maintaining the safety of the global airspace system. For those of us who develop these systems, the key is to remain vigilant about the physics of collision avoidance, the fidelity of our models, and the rigor of our testing. Only then can fixed-wing drones truly become routine participants in the sky.