PROCESS : SEPARATION ASSURANCE 30/05/2014 Context Introduction - - PowerPoint PPT Presentation

HYBRID MODELING AND AUTOMATION OF AIR TRAFFIC CONTROLLER DECISION PROCESS : SEPARATION ASSURANCE

30/05/2014

Context

• Introduction
• Dynamic Modeling of the ATC Process
• Aircraft Dynamic
• Model for Flight Management System
• Flight Plan
• Decision Process of ATCO
• Decision Mechanism of Enroute Controller
• Decision Mechanism of Approach Controller
• Hybrid System Modeling of Air Traffic Controller
• ACC Finite State Automaton
• APP Finite State Automaton
• Algorithm for Automated Safety Assurance
• ACC Algorithm
• APP Algorithm
• Implementation and Results
• Conclusions
• Future Works

I. Introduction

Air Traffic Growth

• The system is still reliable but air transport goes on to grow.

[Source: ICAO, Airbus]

Air Traffic Growth From The View of ATCO How do you handle this?

2014 2030

Human Decision Making

• Human decision making:
• complex
• Innovative

but only for smaller problem size!

Automation tools for APP and ACC

• Goal : Automation tools which perform routine separation

provision tasks of controller for two different types of flight modes.

• Arrival/Departure (APP)
• En-route (ACC) operations
• We utilize hybrid automata formalism for both of these flight

modes.

Air Traffic Management Structure

[Taken from Overview of ATC Systems and Processes , Prof. Hamsa Balakrishnan]

Our Focus

The main subject of this study:

Detailed model for

• ne aircraft (> 5 min)

ACC APP

II. Dynamic Model of the ATC Process

II. Dynamic Model of the ATC Process

• We use an aircraft and flight management model to simulate of Air

Traffic Control (ATC) actions.

• based on Lygeros and Glover’s model [Lygeros,2007]
• We revise the model in some aspects as seen in figure:

• A. Aircraft Dynamic
• Point Mass Model (PMM)
• The state variables of the aircraft are:
• the horizontal position (x1 and x2)
• altitude (x3)
• the true airspeed (x4)
• the heading angle (x5)
• the mass of the aircraft (x6)
• The control inputs of the aircraft are:
• the engine thrust (u1)
• the bank angle (u2)
• the flight path angle (u3)

                 

4 5 3 1 4 5 3 2 4 3 3 2 3 4 1 3 6 6 3 4 2 6 1

cos cos sin cos sin sin 2 sin 2

D L

x x u w x x u w x u w C S x x u x g u x x C S x x u x u           

                         

• B. Model for Flight Management System
• The FMS works like a controller.
• FMS model has 8 discrete modes. These discrete modes are:
• flight level (FL)
• way-point index (WP)
• acceleration mode (AM)
• climb mode (CM)
• speed hold mode (SHM)
• flight phase (FP)
• reduced power mode (RPM)
• troposphere mode (TrM)

u = f(x, flight plan, ATC actions)

Finite State Machines for Flight Manegement System

Finite state machine for AM Finite state machine for FL

These modes are defined relative to BADA for determination of control inputs. Detailed information about these modes can be reached at [Lygeros, 2007]

FMS Controller

• FMS controller can be divided into two components:
• First one is vertical and along track motion control with u1

(thrust) and u3 (flight path angle)

• The second one is horizontal motion control with u2 (bank angle)

• C. Flight Plan
• We use a flight plan data set which includes sequence of way-

points, in three dimensions.

• Way-point data come from point profile of ALLFT+ data set [PRISME

Data]

• ALLFT+ :

point profile of ALLFT+ data set:

Progression to FMS Model

• Flight plan is captured from ALLFT+ data
• The speed profile is evaluated from BADA model

[Sample for BADA performance file]

III. Decision Process of ATCO

III. Decision Process of ATCO

• We determine the decision process of ATCOs speaking with real air

traffic controllers and convert these processes to hybrid models.

nlatim

Decision Process of Air Traffic Controller [Garcia-Avello, 1997]

Decision Process of ATCO

• Decision process of Air Traffic Controller:
• evaluates information
• analyses current situation
• monitors flights
• estimates routes
• detects the problem
• finds solution

Decision Process of Air Traffic Controller [Garcia-Avello, 1997]

• A. Decision Mechanism of En Route Controller
• checks flight levels
• Controller checks flight routes. If:
• Controller checks horizontal separation (5 nm) at intersection point and

longitudinal separation (10 nm ) after intersection point. If any separation losses:

• command direct routing, or
• altitude change, or
• delaying motions
• If:
• Controller checks horizontal separation (5 nm). If any separation losses:
• command direct routing, or
• heading angle change

• B. Decision Mechanism of Approach

Controller

• Controller sequences arrival flights relative to the estimated

arrival time. If any separation losses are estimated between arrival flights:

• delaying motions
• If any separation losses are estimated between departure flights:
• command direct routing, or
• delaying motions

• B. Decision Mechanism of Approach Controller
• If any separation losses are estimated between departure flight

and arrival flight:

• command direct routing, or
• delaying motion, or
• horizontal motion at a defined altitude

Horizontal motion

IV. Hybrid System Modeling of Air Traffic Controller

• A. ACC Finite State Automaton
• ACC Finite State Automaton has eight discrete states which

symbolize defined controller actions:

• q0 is initial state which refer to no action.
• q1 denotes direct routing for first aircraft
• q2 denotes altitude change for first flight
• q3 denote delaying motion for second flight with reducing of speed
• q4 denote delaying motion for second flight with vector for spacing
• q5 denotes altitude change for second flight
• q6 denotes direct routing for both of them
• q7 denotes bank angle for both of them

• A. ACC Finite State Automaton
• The logic definitions for transition functions and helper functions

are:

 

   

 

   

 

         

1 1 1 2 4 1 2 4 3 2 1 2 4 3 1 2 4 3 1 2 4 3 5 4 1 4 5 5 1 4 2 6 1 4 2 5 7 5 8 1 4 5

, , 1, and

• r a

a is a not a a is e a a a a a a a a a a a a e a a a a a a a a a e a a a a a a e a a a a e a a a a e a a a a a e a a e a a a a                                                        

B. APP Finite State Automaton

• APP Finite State Automaton has seven

discrete states:

• q0 is initial state which is defined as no

action

• q1 denotes direct routing for second

flight (departure flight)

• q2 denotes delaying motion for second

flight which is applied with reducing of speed (ROCD)

• q3 denotes horizontal motion for

second flight at a defined altitude. In q3, departure flight are climbed to a defined altitude, moved along track and climbed to original route for separation with arrival flight

q3

• B. APP Finite State Automaton
• q4 denotes delaying motion for first flight
• q5 denote delaying motion for second flight with vector for spacing
• q6 denote delaying motion for second flight with reducing of speed

• B. APP Finite State Automaton
• The relation between transition functions and helper functions

are:

    



 

  



 

       

1 2 3 2 2 2 3 6 3 1 2 3 4 1 2 5 1 2 3 5 6 5 4 6 4 5 7 7 2 2 3 6 8 1 2 1 2 3 6 9 4 7

, , 1, and

• r a

a is a not a a is e a a a e a a a a a a e a a a e a a a a a a a a e a e a a a e a a a a a a e a a a a a a e a a                                            

V. Algorithm for Automated Safety Assurance

A. ACC Algorithm

ACC Controller Algorithm Flight plans of all enroute flights and current state variables Controller actions and new flight routes of all enroute flights a new aircraft comes to sector Algorithm 1: input :

• utput :

1 if then 2 Check separation of all flights in sector any unseparated flight exist all unseparated flights in sector flight1 to most old aircraft in unsepareted flights all unseparated flights with flight1 3 if then 4 for to | do 5 Set 6 for to | do flight2 to most close unseparated flight to flight1 controller action from ACC Automata for flight1 and flight2 new flight1 and new flight2routes to new flight routes 7 Set 8 Generate 9 Set

• The algorithm generalizes ACC Automaton to multi-flight handling

within the sector.

B. APP Algorithm

• Sequencing of arrival flights
• Separating arrival-arrival

conflict with APP Automaton

• Separating arrival-departure

conflict with APP Automaton

• Separating departure-

departure conflict with APP Automaton

• C. Computational Complexity of Algorithms
• Algorithms have two parts:
• conflict detection
• separation assurance

time_conflict detection number of aircraft x number of way-points

• f each aircraft

time_separation assurance number of aircrafts which have separation losses x number of way-points of each aircraft which have separation losses

 

Time of conflict detection

20 40 60 80 100 120 140 2 4 6 8 10 12 14 16 Number of Flights in a Fixed Level Computation Time of Conflict Dedection (s) 6 way-points 10 way-points 14 way-points 18 way-points 10 20 30 40 50 60 0.2 0.4 0.6 0.8 1 1.2 1.4 Number of Flights in arrival Computation Time of Conflict Detection (s) 9 way-points 12 way-points 15 way-points

CD for Enroute CD for Approach

Time of separation assurance

10 20 30 40 50 60 70 5 10 15 20 25 Number of Flight in a Fixed Level Computation Time of Separation Assurance (s) 5 10 15 20 25 30 35 40 0.5 1 1.5 2 2.5 3 3.5 4 Number of Flight in arrival (number of departures is same) Computational Time of Separation Assurance (s)

SA for Enroute SA for Approach

VI. Implementations and Results

A. Implementation for Enroute

• We used real flight data from 1 March 2011 for implementations.

We use two different sets:

• First data set includes all flight trajectories between 11:00 - 11:15 in

Istanbul ACC

• Second data set includes all flight trajectories between 11:00 - 13:00

• A. Implementation for Enroute
• In first set, 18 flights have been seen in 15 minutes within the

sector

• Two of them have lost separation
• ACC Controller Algorithm solved this separation lost by giving

vector command to an aircraft

• A. Implementation for Enroute
• In second set, 102 flights have been seen in 120 minutes within

the sector

• 13 of them have lost separation
• ACC Controller Algorithm solved these separation lost by

intervening to 7 aircrafts:

• 2 altitude changes
• 1 direct routing
• 1 reducing of speed
• 3 VFS

B. Implementation for Approach

• We used real flight data from 1 March 2011 for implementations.

We use two different sets:

• First data set includes all flight trajectories between 18:00 - 20:00 for

APP sector which including to Sabiha Gokcen Airport

• Second data set includes all flight trajectories between 18:00 - 20:00

for APP sector which including to Ataturk Airport

• B. Implementation for Approach
• In first set, 13 flights have been seen in arrival and 19 flights have been

seen in departure between 18:00 - 20:00 for Sabiha Gökçen Airport

• 2 of arrival flights and 2 of departure flights have lost separation
• APP Controller Algorithm solved these separation lost by

intervening to 2 aircrafts:

• speed change for 1 arrival flight
• VFS for 1 departure flight

Violet ----- departure flight Blue ----- arrival flight Yellow ----- intervened flight

• B. Implementation for Approach
• In second set, 24 flights are appeared in arrival and 31 flights

appear in departure from 18:00 to 20:00. 2 of arrival flights and 6

• f departure flights will have loss of separation.
• In second set, 24 flights have been seen in arrival and 31 flights have

been seen in departure between 18:00 - 20:00 for Atatürk Airport

• 2 of arrival flights and 6 of departure flights have lost separation

• B. Implementation for Approach
• APP Controller Algorithm solved these separation lost by

intervening to 2 aircrafts:

• speed change for 1 arrival flight
• reducing of speed for 1 departure flight
• VFS for 2 departure flight

Violet ----- departure flight Blue ----- arrival flight Yellow ----- intervened flight

Conclusions

• We have presented two different hybrid system models for the

decision process of Air Traffic Controller in en route and approach

• peration.
• By using these models, we have designed an automation algorithm

for achieving separation assurance.

• By using real traffic data, we have shown that the algorithm can

detect conflicts and recommend solutions at seconds.

• The workload of controllers can be reduced with proposed

automation tools and capacity of current system can be enhanced.

Future Works: Integration to Flight Deck Testbed

Future Works: Enhancing with flow management

• Optimization algorithms for flow management

Thank You !