Universal Deconfliction Model for UTM

The somewhat mystical "deconfliction" term often bantered about in UAS Traffic Management (UTM) discussions is sorely in need of more definitional rigor. This is evident after spending countless hours researching, studying and attempting to apply the concept to the activity of small drones and most recently, to urban air mobility. We have taken up the effort to help bring more clarity, uniformity and completeness to the subject and this document presents a Universal Deconfliction Model (UDM) as a result.

This model leverages the work of ICAO, FAA, NASA and others in industry and academia as cited in the references section at the end of this document. It represents a distillation, normalization, synthesis and extension of the best insights on conflict mitigation striped across an abundance of relevant material including several publications and presentations that deal with the deconfliction topic. Unfortunately, this body of work is highly fragmented, inconsistent and generally incomplete. The majority of it is also tailored to manned aviation or large drones equipped to operate under instrument flight rules, rather than small drones in UTM airspace.

This document is meant to help remedy this situation for UTM and stands in direct contrast to the kinds of material you typically see in both civil and military aviation circles on subjects of this nature. You know the stuff we are talking about - highly detailed and complex to a fault, full of whiz-bang graphics and eye candy, and crammed with disjointed concepts and half-baked thoughts. Throw in a lack of standard nomenclature and lexicon and it is difficult to extract much in the way of a coherent framework from these contributions as a whole.

We have attempted to avoid these pitfalls with the hope of promoting an easily understood and shareable model that can facilitate conflict management within UTM. In order to appropriately set expectations, please note this is a primer targeted to the "for dummies" audience, and not a doctorial thesis. We have far too much of the latter and very little of the former, providing impetus for this effort.

It should be noted the discussion presented in this document applies to UTM as well any digital variant often described as a “UTM inspired” manifestation of air traffic management. This includes UAM traffic management capabilities envisioned to support emerging use cases characterized by passenger and cargo drones.

Definition

Before we start, a proper definition of deconfliction is probably in order. The term itself is defined in military and engineering circles to describe a situation that can be actively managed, and refers to a process for avoiding mutual interference or outright hazards among entities or elements in the same domain.

In the aviation world it commonly refers to the need to recognize and mitigate a loss of safe separation between aircraft and other potential sources of conflict including terrain, obstacles and weather. The scope of deconfliction in this domain extends across the full operational horizon from pre-flight planning to in-flight activity to recognize and respond to a potential safety threat in a timely fashion. The former is typically referred to as strategic deconfliction and the latter as tactical deconfliction.

Preface it to say the “D” word sounds tacky and somewhat edgy. We prefer "conflict mitigation" which has a more elegant feel to it. Regardless, deconfliction is a de facto industry term used here to describe a comprehensive and stage oriented process that addresses a full range of conflict elements faced by an aircraft cooperatively operating under the auspices of UTM.

Universal Scope

Several aspects collectively define and contribute to the universal scope of the proposed deconfliction model for UTM;

• End-to-end stage oriented process
• Comprehensive conflict mitigation
• Normalized subject matter foundation
• Harmonized nomenclature and lexicon
• Extensible to new entrants in aviation
• Technology architecture agnostic
• Reusable on multiple levels.

The proposed model prescribes an end-to-end process that fully addresses all stages of deconfliction. This stands in contrast to other contributions that are either incomplete or avoid a process centric and how-to discussion in favor of simply identifying "layers" of a conflict management onion. This model also addresses a comprehensive set of conflict scenarios that go well beyond aircraft contending for the same airspace, which tends to be the extent of many contributions that simply ignore other equally hazardous situations.

Normalization of contributions that comprise the subject matter is accomplished by distilling and synthesizing this body of work into a common and broadly applicable framework serving as the foundation for the proposed model. This includes harmonization and standardization of sometimes widely varying nomenclature and lexicon, opting to preserve the most common and meaningful usage but to also incorporate nuances where warranted.

New entrants in the aviation industry are uniquely supported by the proposed model including small unmanned aircraft systems (sUAS) and urban air mobility (UAM) platforms envisioned for the future. In addition, the model does not impose any particular technology architecture on UTM and is a malleable construct that is reusable on several levels, including the process view as well as related system design and software derivatives.

So why the emphasis on a universal model in the first place? In short, it promotes abstraction of the problem domain that facilitates reuse of system artifacts to improve quality, maintainability and time to market. Deconfiction is a major function in UTM and these traffic management services will be a highly automated digital incarnation in contrast to analogous human centric services provisioned by ATM and ATC today. A universal model aids the design, development and implementation of the software based ecosystem that both defines and comprises UTM.

The following illustration connotes the idea of reusability in this context. We urge you to not get wrapped up in the nuances of the model or its derivatives at this point, since this is explained in detail in the sections that follow. Instead, the key notion here is the Universal Deconfliction Model provides a general framework that can be cloned and tweaked for various use cases with each representing a different source of conflict in UTM.

For some final perspective, universal applicability was not the original scope behind the creation of this model. Like many, we initially started with a very narrow focus on deconfliction of aircraft in UTM airspace, being most concerned about the potential for a mid-air collision with BVLOS operations. It then became apparent this use case could be abstracted into a more general and universal model that readily applies to other sources of conflict such as weather, terrain and obstacles which can be managed under the same stage oriented process model.

Comprehensive Conflict Mitigation

With respect to a universal model, it is important to recognize the scope of deconfliction in UTM applies to more than just trying to prevent a mid-air collision with another aircraft. Conflict can take many forms as listed below, all of which can contribute to a loss of safe separation that needs to be properly mitigated.

You can refer to the sources enumerated here as different conflict elements that may need to be dealt with in each stage of deconfliction under UTM. Some of these represent an outright hazard and others directly interfere with cooperative operations envisioned in UTM.

Please note the "starburst" sandwiched by the diametrically opposing arrows in the illustration serves as a generic symbol for any or all of the potential sources of conflict, depending on the context.

Ownship refers to an aircraft operating under the auspices of UTM and using a range of related automated traffic management services and capabilities. A rogue or otherwise non-cooperative aircraft bulleted at the top of the left hand column is just one example of a conflict element that represents a potential threat to ownship.

Some of these conflict elements are concrete in nature with very well defined physical characteristics including obstacles such as a tower crane commonly used in construction. These may be the easiest to deal with given their visibility and recognizable physical dimensions. Others are non-visual or amorphous and can present a greater challenge for conflict mitigation. An example of this may be the threat of hazardous weather including wind gusts that make operation difficult. The proposed Universal Deconfliction Model can support both categories equally well.

End-to-End Stage Oriented Process

A stage oriented process is instrumental to the proposed Universal Deconfliction Model since it addresses all aspects of conflict mitigation in UTM, from pre-flight planning on a strategic level to in-flight or tactically oriented deconfliction. It also covers a broad spectrum of conflict elements commonly encountered by drones and UAM aircraft in low-altitude airspace.

The approach used here provides a conflict mitigation framework specifically adapted to UTM. It includes four stages that successively address Strategic Deconfliction, Separation Management, Self-Separation, and Collision Avoidance or Encounter Avoidance associated with various types of conflict faced by cooperative aircraft using UTM services.

This implies a temporal sequencing of these functions although progression through some stages may not occur in practice for a variety of reasons. Communication issues could temporarily disrupt the provision of separation management services in Stage 2 for example, calling for the pilot or operator to assume separation responsibility and self-separate in Stage 3 in response.

An important nuance of this stage orientation is the notion of “preemptive deconfliction” which results in a successive reduction in both the probability and degree of potential conflict as you progress through the various stages. Thus, Strategic Deconfliction on a pre-flight basis in Stage 1 is instrumental to reducing the need for tactical or in-flight conflict mitigation capabilities in Stages 2 - 4. The same applies to Separation Management in Stage 2, which similarly reduces the need for self administered deconfliction by ownship in Stages 3 - 4, although it does not eliminate this requirement.

Stage 1 - Strategic Deconfliction aims to preemptively resolve conflicts within and between the operation plans of various airspace users and other known constraints prior to launch. On a practical level, this refers to the elimination of intersecting operation volumes or trajectories on a geospatial and temporal basis, and the honoring of both static and dynamic constraints such as no-fly zones, obstacles, weather and other recognized hazards.

Stage 2 - Separation Management is an automated UTM capability analogous to in-flight separation services provided to manned aviation by a human controller in ATC. In this case, the operator of ownship subscribes to these hosted services in order to benefit from early recognition of a potential conflict and safe and efficient resolution enabled by a holistic view of the airspace and its state, including current and planned operations.

Stage 3 - Self-Separation relies on the pilot or operator to see and avoid a hazard and remain well clear of it. The mechanics of this can be manual, assisted or automated but it marks a clear transition from reliance on and use of separately provisioned tactical deconfliction services under UTM to a self administered mode by ownship. This delegation of separation responsibility occurs in a situation with an increased level of risk that is best managed by the pilot or operator.

Stage 4 - Collision Avoidance includes a maneuver of last resort to prevent a mid-air collision or other type of encounter depending on the use case. This can also be invoked by the pilot on a manual basis or it can be automatically executed, but requires immediate attention and action to mitigate a situation where the potential for collision is now unacceptably high. Note this “collision” nomenclature is more attuned to common physical hazards such as other aircraft, terrain or obstacles. A more generic alternative is Stage 4 - Encounter Avoidance which is applicable to a universal set of conflict elements including others of a non-visual or amorphous nature such as wind gusts, degraded GPS, or congestion.

The four stages that make up the aircraft deconfliction use case are shown in the mosaic below, beginning with Stage 1 (Strategic Deconfliction) in the upper left and flowing clockwise around the horn.

In a situation where the conflict source is physical in nature such as another intruder aircraft that presents a threat to ownship, localized or on-board Detect and Avoid (DAA) capabilities become an important facilitator of deconfliction. DAA by definition spans both Stage 3 (Self-Separation) and Stage 4 (Collision Avoidance) under the proposed model and can be manual, assisted or automated in nature.

Universal Deconfliction Model (UDM)

The following illustration includes the proposed Universal Deconfliction Model for UTM. A detailed examination of specific use cases such as deconfliction of aircraft, weather and obstacles that are derived from this universal model is provided in subsequent sections. The potential for substantial reuse afforded by the general-to-specific relationship between the universal model presented here and the specialized use cases that follow should ring clear upon visual inspection and consideration.

Ownship is the aircraft operating under UTM and in the universal model is represented by a small unmanned aircraft system (sUAS) or drone. However, this could easily be a UAM platform such as an air taxi envisioned for the future, and the integrity of the universal model still holds.

A close examination of this model shows Stages 1 - 4 progressively arranged from left to right across the top half of the illustration. Each stage incorporates the generic conflict symbol consisting of the starburst sandwiched by opposing arrows, implying applicability to any and all of the various conflict elements previously enumerated.

Stage 1 (Strategic Deconfliction) consists of planning related mitigation activities on a pre-flight basis, performed before activation or launch and is recognized as a Strategic Deconfliction stage without strict time constraints. Stages 2 – 4 represent in-flight or Tactical Deconfliction stages where the time reference denoted in red under each infers the hypothetical window available to recognize and resolve the conflict before it escalates to the next stage. In practice this time will be individually prescribed for and vary by the performance capabilities and equipage of ownship as well as consideration of the use case for each conflict source such as an intruder aircraft, hazardous weather, or obstacles.

Note that the ICAO definition of Separation Provision which includes Stages 2 - 3 is referenced in the model for consistency in nomenclature. However, deconfliction typically progresses through both stages in this model whereas the ICAO definition intimates the designation of one over the other in practice.

In addition, Detect and Avoid (DAA) combines Self-Separation and Collision or Encounter Avoidance under stages 3 and 4 as previously noted. This is consistent with the recommendations of a UAS working group commissioned by the FAA to define the scope of DAA. They determined it to be a multi-faceted and two stage concept that approximates the “see and avoid” responsibility in manned aviation, which requires the pilot to maintain proper vigilance to steer well clear of other aircraft and avoid potential hazards that represent a danger to safe operation.   

Ownship is encased by different airspace volumes and thresholds represented as virtual three dimensional cylinders in the lower left, and in an equivalent top view rendering for Stages 2 - 4 striped across the upper portion of the illustration. These volumes and thresholds move with the aircraft and allow for the provision of different services on a conflict proximity basis. This is different from the concept presented in many contributions where similar cylinders encase the intruder or target aircraft, in addition to or in lieu of the same for ownship. The size of these cylinders is automatically defined by UTM and is based on ownship equipage and performance characteristics as well as the nature of the conflict element.

The Tactical Deconfliction Volume (TDV) is the first cylinder breached by a conflict element such as an intruder aircraft. This triggers the provision of automated separation management services by UTM on behalf of ownship under Stage 2 (Separation Management). In this case this digital capability has minutes to recognize the potential conflict, alert ownship, and determine and communicate an optimal resolution strategy for ownship and other affected cooperative aircraft to execute in order to deconflict the situation and avoid loss of separation.

The next threshold subject to a breach is the Self-Separation Threshold (SST) which demarks the beginning of Stage 3 (Self-Separation). In this stage the pilot or operator unilaterally assumes control to assess the emerging threat posed by the conflict element and decide what action to take to remain well clear. The risk has now risen to an unacceptable level with perhaps a minute or so to resolve the situation, allowing some time for a rationalized maneuver but not for an exhaustive consideration of the safety and efficiency issues associated with it. The objective here is to execute a Well Clear Maneuver (WCM) to maintain safe separation and halt progression to stage 4 and the need for a maneuver of last resort.

As previously noted, this activity can either be automated or manually invoked by the pilot or operator in an effort to see and avoid the conflict element.

Stage 4 (Encounter Avoidance) is triggered by a breach of the Encounter Avoidance Threshold (EAT) surrounding ownship and represents just seconds for ownship to initiate a maneuver of last resort in order to escape the consequences associated with the conflict element. The technological implementation at this stage is analogous to TCAS and more recently to ACAS in the manned aviation regime.

Please note the nomenclature at this stage may be modified when the model is applied in practice, in order to best fit the particular use case. More specifically, the “encounter” term is replaced with “collision” where the conflict element clearly implies a physical collision is at hand, such as with a rogue aircraft that poses a threat to ownship.

The Encounter Volume (EV) surrounding ownship defines the approximate space occupied by the aircraft, where a breach is almost certain to produce a direct encounter with the conflict element including the negative ramifications associated with it. The size of the EV closely conforms to but varies by the actual dimensions of ownship, adjusted slightly for navigational error. Due to the need for an instantaneous response this stage typically leverages on-board systems to quickly recognize the situation and immediately execute an Encounter Avoidance Maneuver (EAM) to prevent the likely mayhem associated with a breach of the EV.

In summary, the Universal Deconfliction Model provides a framework for adaptation to specific use cases including deconfliction of aircraft, weather, obstacles and other outright hazards or interference to safe operations. The following section addresses deconfliction of aircraft and is meant to illustrate how this universal model can be applied to a specific use case and by analogy, to the other sources of conflict in a UTM context as well.

Deconfliction of Aircraft in UTM

This use case addresses one of the most pressing issues in enabling drones and urban air mobility vehicles to safely share the same airspace with manned aviation. That is, how to prevent a mid-air collision between manned and unmanned aircraft flying in close quarters. Since it is the most common scenario discussed with deconfliction in UTM, it is used to show how the universal model can be leveraged for a specific purpose. The illustration provided below reflects this adaptation for deconfliction of aircraft where a non-cooperative airplane initially intrudes on ownship and continues to progress to a threat status.

The similarity of this use case to the Universal Deconfliction Model as previously presented should be obvious, with only a few very subtle changes to the nomenclature and lexicon, mostly in Stage 4 which is referred to as Collision Avoidance in this context.

Ownship in this use case is a small unmanned aircraft system (sUAS) or drone which is operating under UTM. The generic starburst conflict symbol is replaced in this context with a fixed wing airplane of the type typically flown by general aviation (GA) cohorts. The GA aircraft is non-cooperative meaning it does not engage with UTM or otherwise afford any electronic conspicuity.

In this scenario, the deconfliction process starts on a pre-flight basis under Stage 1 (Strategic Deconfliction). The operation plans filed for ownship and other cooperative UTM airspace users are reviewed to identify and eliminate intersecting operation volumes or flight trajectories on a 4D basis (three geospatial dimensions along with time). Other known conflict elements are also mitigated at the planning level to minimize the probability and frequency of downstream conflicts arising in-flight under Stages 2 - 4. There are no strict time limits for this stage except for the need to conclude this effort prior to activation of the operation plan by the pilot or operator and the launch of ownship.

Stage 2 (Separation Management) marks the start of tactical or in-flight deconfliction activities and is triggered when the Tactical Deconfliction Volume (TDV) surrounding ownship is breached by the GA aircraft intruding on its operation. This stage recognizes a potential conflict in the making but still affords minutes for a highly optimized resolution that ensures the safety and efficiency of the airspace, since the risk is relatively low at this point. The separation management services provisioned on behalf of ownship are automated in UTM and include recognition of the breach, timely posting of a traffic alert, and the optimal resolution and communication of same to ownship and all cooperative aircraft that are affected. This effort is facilitated by a full view of the overall state of the airspace commanded by UTM, including the status of current and planned operations.

The situation may continue to escalate despite previous deconfliction efforts. A subsequent breach of the Self-Separation Threshold (SST) surrounding ownship marks the start of Stage 3 (Self-Separation) where the risk has risen to an unacceptable level, and necessitates a resolution within about a minute to avoid a potential collision. The pilot or operator now makes a unilateral decision and maneuver to remain well clear of the recognized threat posed by the GA aircraft. This may be executed manually or assisted by an automated airborne capability of ownship, but it no longer relies on the provision of separation services by UTM. Due to the time constraint, it does not allow for a fully optimized resolution but looks to execute a reasonably safe and timely maneuver to escape a collision.

In the event the GA aircraft ultimately breaches the Collision Avoidance Threshold (CAT) of ownship to trigger Stage 4 (Collision Avoidance), an immediate and drastic Collision Avoidance Maneuver (CAM) by ownship is necessary with only seconds available to prevent a compromise of its Collision Volume (CV) and corresponding mid-air collision. The CAM can again be manually executed by the pilot or it can be an automated maneuver enabled by an on-board capability but it needs to be as simple as possible to minimize the decision process and reaction time involved, with the only goal to avoid a mid-air collision.

Deconfliction of UAM Aircraft in Next Generation UTM

Urban air mobility (UAM) creates special challenges for traditional air traffic management in urban airspace. One school of thought says the industry will need an enhanced version of UTM to support passenger and cargo drones at scale. This is frequently referred to as a “UTM inspired” incarnation and like UTM is a digital and networked solution based on a service oriented architecture which leverages federated services provided by a number of different commercial entities. Some refer to this next generation as UAM traffic management (UAMTM) to distinguish it from UTM which focuses on small unmanned aircraft systems or sUAS.

A close examination of the illustration below shows it essentially unchanged from the previous section which deals with the deconfliction of sUAS operating in low-altitude airspace under the conventional manifestation of UTM. This implies the deconfliction process for UAM aircraft is the same whether we are dealing with small drones or alternatively, with passenger or cargo drones as is the case with UAM.

On a functional level, this next generation of urban air traffic management will build upon many of the same strategic and tactical deconfliction capabilities in UTM but with some noteworthy changes as follows.

Strategic Deconfliction in Stage 1 will be based on a four-dimensional trajectory (4DT) including the three spatial dimensions plus time, in contrast to the airspace volume reservation model for sUAS under conventional UTM. Separation Management in Stage 2 will need to work seamlessly with flow control in and out of vertiports which will be a unique contention point and resource constraint under UAM. More sophisticated and enhanced Tactical Deconfliction capabilities covering Stages 2 - 4 will be required to support a heterogeneous mix of UAM aircraft with widely varying performance capabilities and equipage. Finally, more robust surveillance capabilities in support of Separation Provision in Stages 2 - 3 will be required to provide more timely and precise position reporting in support of precision navigation for UAM with a smaller margin for error.

Please note the designated times that serve as a window to either resolve or escape collision at each stage of the deconfliction process are most representative of the early phase of UAM adoption where activity is not expected to be extensive. These times will need to be reduced to accommodate higher density and tempo operations under this use case.

Again, the need to implement preemptive deconfliction across the four stages included here is based on the notion the probability and degree of potential conflict can be reduced and perhaps eliminated as you progress from stage to stage. This is particularly important with respect to mitigating the need for a Collision Avoidance Maneuver in Stage 4 which can create cascading secondary effects in the airspace that may be equally dangerous and highly inefficient to mitigate.

Other Deconfliction Use Cases

The previous section on the deconfliction of aircraft in UTM was meant to illustrate how the Universal Deconfliction Model can be readily adapted to one of the most prevalent conflict mitigation issues associated with different aircraft trying to operate in the same airspace. This section strives to reinforce the general-to-specific relationship between the universal model and its application to other sources of conflict or use cases, including obstacles and weather. Additional conflict elements beyond these and which were previously identified are not addressed in any further detail in this document, but should follow the same reuse paradigm.

With respect to both obstacles and weather in UTM, these represent conflict elements with different characteristics that place them at opposite ends of the mitigation spectrum. However, they can be effectively addressed with the same stage oriented deconfliction process and conflict proximity triggers conveyed by the universal model.

An obstacle such as a construction crane has well defined physical characteristics and represents an outright hazard with catastrophic consequences in a collision scenario. Hazardous weather such as freezing rain on the other hand, is amorphous and in constant motion. Although an encounter with the latter does not necessarily cause immediate physical mayhem it could easily contribute to this and should be treated like any other potential hazard.

The illustrations for these use cases should look very similar to the deconfliction of aircraft as well as the universal model presented in the previous sections. Again, the implication here is that reuse is possible at the process level and in system design and implementation efforts across a comprehensive set of conflict elements, to speed delivery and improve system quality and maintenance.

Highly Conflicted Airspace of the Future

UTM will need to safely deconflict airspace filled with drones and a variety of other new entrants as well as traditional sources of conflict including weather, terrain, obstacles and other airborne threats. Maintaining operational efficiency will be a close secondary consideration for deconfliction under UTM, with the goal to avoid significant change to planned and active operations where possible. This calls for a minimally invasive approach to conflict resolution and a focus on the most simple mitigation strategy to address a particular conflict scenario.

This will be easier said than done but will be mandatory for successful adoption of a plethora of new types of aircraft, modes of operation and business models in aviation.

References

This section identifies some of the major sources of concepts, ideas, nomenclature and lexicon referenced in the generation of this document. This includes the following artifacts along with a multitude of other similarly relevant but less pertinent contributions in the distillation, normalization, synthesis and extension of this entire body of work in order to produce a Universal Deconfliction Model for UTM. In summary, this model leverages a wealth of knowledge and insight to inherently address the nuances of deconfliction in UTM and provide a harmonized reference for industry guidance.

1. Global Air Traffic Management Operational Concept - Doc 9854, ICAO, 2005
2. Unmanned Aircraft Systems (UAS) Traffic Management (UTM) Concept of Operations, Federal Aviation Administration, May 18, 2018
3. Unmanned Aircraft System Traffic Management (UTM) Concept of Operations, NASA Ames Research Center, June 2016
4. Blueprint for the Sky: The Roadmap for the Safe Integration of Autonomous Aircraft, A3 by Airbus, August, 2018
5. Sense and Avoid (SAA) for Unmanned Aircraft Systems (UAS), Federal Aviation Administration, October 2009
6. Sensor and Tracker Requirements Development for Sense and Avoid Systems for Unmanned Aerial Vehicles, American Institute of Aeronautics and Astronautics, August 2018
7. Analysis of Autonomous Deconfliction in Unmanned Aircraft Systems for Testing and Evaluation, 2009 IEEE Aerospace Conference
8. Manual on Remotely Piloted Aircraft Systems - Doc 10019, ICAO, 2015

*************************************************************************

v20191105

© 2018 The Fresnel Group