Hess - UCDavis by xiangpeng

VIEWS: 8 PAGES: 21

									SAE/IEEE Aerospace Control and Guidance
          Systems Committee
                  Meeting 102
             Grand Island, New York
               Oct. 15 – 17, 2008

                      Ron Hess
   Dept. of Mechanical and Aeronautical Engineering
               University of California
                      Davis, CA
                                Outline

• University of California Davis Aero Program

• Analytical Approach to Assessing Flight Simulator Fidelity

• Modeling Pilot Adaptation to Sudden Changes in Vehicle Dynamics

• Modeling Human Pilot Controlling Rotorcraft with Time-Varying
  Dynamics



                 Sponsor: NASA Subsonic Rotary Wing Project;
                          Technical Manager: Dr. Barbara Sweet
                      UCD Aero Program
                      25 Year Celebration
• UC Davis Aeronautical Science and Engineering Program Celebrating 25
  years since initial accreditation by ABET

• First accredited Aeronautical/Aerospace Program in the Nine Campus UC
  System
                          UC Davis Aero Faculty

  Jean Jacques Chattot (Dept. Chair)   Valeria LaSaponara
  Roger Davis                          Nesrin Sarigul-Klijn
  Mohamed Hafez                        Bruce White (new Dean of Eng.)
  Ron Hess                             Case van Dam
  Sanjay Joshi
Robert Mondavi Food and Wine Institute
        University of California
                 Davis
Robert Mondavi Center for Performing Arts
        University of California
                 Davis
        Analytical Assessment of Flight Simulator Fidelity
•Pilot Model Developed That Includes
 –Visual feedback with degraded cues            - Proprioceptive feedback
 –Vestibular feedback                           - Task interference
 –Variable skill levels


•Aimed Toward Assessing Training Simulator Fidelity
“We suggest, then, that fidelity is the specific quality of a simulator that permits the
skilled pilot to perform a given task in the same way that it is performed in the actual
aircraft. Execution …is simply the closure of all loops made necessary by both the task
requirements and the dynamics of the vehicle and subject to the information available.”
          - Heffley, R. K., et al, “Determination of Motion and Visual System
             Requirements for Flight Training Simulators,” U.S. Army Research
             for the Behavioral and Social Sciences, TR 546, Aug. 1981.
       Fidelity Example: Small Rotorcraft – BO-105




• Task: Reposition task (4 control axes) with atmospheric turbulence
• Flight Condition: near hover
• Simulator “limitations” – 4 scenarios
     - no motion
     - limited motion
     - limited motion + reduced visual cue quality
     - limited motion + reduced visual cue quality + time delay in sim
            Fidelity Example: Small Rotorcraft – BO-105




                     pilot model for longitudinal control loops




pilot/vehicle computer simulation model     power in proprioceptive feedback signal
Fidelity Example: Small Rotorcraft – BO-105
              Fidelity Metrics
          (larger values imply poorer fidelity)
•   no-motion
      FM = pitch + roll + vertical position + heading
          = 1.36 + 2.39 + 0.36 + 0.837 = 4.95
•   limited-motion
      FM = pitch + roll + vertical position + heading
          = 0.4 + 0.7 +.05 + 0.15 =1.3
•   limited-motion + reduced visual quality
      FM = pitch + roll + vertical position + heading
          = 0.89 + 1.28 + 0.22 + 0.62 = 3.01
•   limited-motion + reduced visual quality + time delay
      FM = pitch + roll + vertical position + heading
          = 0.98 + 2.04 + 0.208 + 0.07 = 3.3
                                                       Fidelity Example: Large Rotorcraft – CH-53D
                                                                                                                                                             -6
                                                                                                                                                         x 10
                                                                                                                                                   3.5




                                                                                                        P() for roll-loop proprioceptive signal
                                                                                                                                                    3


                                                                                                                                                   2.5


                                                                                                                                                    2                                                   no motion
                                                                                                                                                                              motion

                                                                                                                                                   1.5


                                                                                                                                                    1


                                                                                                                                                   0.5


                                                                                                                                                    0
                                                                                                                                                         0        0.2   0.4   0.6   0.8     1     1.2   1.4   1.6   1.8   2
                                                                                                                                                                                    frequency rad/sec



                                                                                                     power in proprioceptive feedback signal
                                  250

                                  200                          x-dot            10 x 
                                                                                                    Fidelity metric calculation is independent
                                  150                                                                      of time-variant task demands
  x-dot, 10  , 10 ft/sec, deg




                                  100

                                   50
                                                                                                   FM = pitch-loop contribution + roll-loop
                                    0
                                                                                                               contribution + vertical velocity-
                                   -50                                10 x 

                                  -100
                                                                                                               loop contribution + heading-rate
                                  -150
                                                                                                               loop contribution
                                  -200
                                         0   10   20   30        40            50        60   70
                                                            time sec



accel/decel task – time varying pilot model                                                           = 0.0148 + 0.02 + 0.107 + 0.0218 = 0.164
         hover - 110 kts - hover
                    Modeling Pilot Adaptation to
                 Sudden Changes in Vehicle Dynamics




                   Adaptive Pilot Model – Single Axis Tasks

                        Four criteria for model adaptation
•   signals must be easily sensed by pilot
•   adaptation completed in 5 sec or less
•   logic in adaptation must be predicated upon information available to pilot
•   post-adapted pilot models must follow dictates of crossover model of human
       Modeling Pilot Adaptation to
    Sudden Changes in Vehicle Dynamics
                     (single-axis task)




Pilot model adapting to suddenly changing vehicle dynamics
                with pulsive commands C(t)
   Modeling Pilot Adaptation to
Sudden Changes in Vehicle Dynamics
(multi-axis task with control cross-coupling)



                                                   1         e 0.025s
                                      Yc1 (s)           
                                                s(s  4)     s 2 (s  4)




                                                       1
                                       Yc 2 (s)             
                                                    s(s  4)
                                                    e  0.025s
                                                s(s  0.5)(s  4)
           Modeling Pilot Adaptation to
        Sudden Changes in Vehicle Dynamics
            (multi-axis task with control cross-coupling)


             1         e 0.025s                     1           e 0.025s
Yc1 (s)                              Yc 2 (s)           
          s(s  4)     s 2 (s  4)                s(s  4)   s(s  0.5)(s  4)




     Pilot model adapting to suddenly changing vehicle dynamics
               with random-appearing commands C(t)
           Modeling Human Pilot Controlling Rotorcraft
                  with Time-Varying Dynamics

                           Pilot Model




      with Ypf
configured properly
         
    UM  M
       Modeling Human Pilot Controlling Rotorcraft
              with Time-Varying Dynamics




From Yc = 1/s to Yc = 25/(s2+6s +25)
                                       cue to pilot that dynamics have changed
                                 Modeling Human Pilot Controlling Rotorcraft
                                        with Time-Varying Dynamics




                                                                High –fidelity model of Army RASCAL

                                        Bode Diagram


                        20



                                                           -20 dB/dec
      Magnitude (dB)




                          0




                        -20
                          0


                       -100
Phase (deg)




                       -200


                       -300


                       -400
                            -1            0                              1
                          10         Frequency (rad/sec)
                                        10                              10




 Pilot model/vehicle open-loop                                                   Pilot/vehicle open-loop transfer function
        transfer function                                                              from laboratory tracking task
                           Modeling Human Pilot Controlling Rotorcraft
                                  with Time-Varying Dynamics

              pitch and roll SCASs changing from RC/ATTH to ATTC/ATTH over 10
                                sec with time-varying pilot model
              -3
           x 10                                                                              1
      8
                                                                                                                                   pilot model
                                                                                           0.8                                     transition
      6                                                                                                y
                                                                                           0.6
                                                                                                                             h            x
      4
                                                                                           0.4


      2                                                                                    0.2

                                                                                             0
 UM




      0
                                                                                           -0.2
      -2
                                                              violation of normal          -0.4
                                                              operating area
      -4                                                                                   -0.6

                                                                                           -0.8
                                                                                                           SCAS
      -6                                                                                                   transition
                                                                                            -1
      -8                                                                                          0   10   20           30   40   50      60     70   80
      -0.06        -0.04      -0.02         0          0.02          0.04           0.06
                            visually-sensed pitch rate rad/sec



cue to pilot that SCAS is changing                                                           pilot/vehicle tracking performance
                                                                                                with time-varying pilot model
         Modeling Human Pilot Controlling Rotorcraft
                with Time-Varying Dynamics

                    Predicting Handling Qualities Levels

                                           5

                                          4.5

                                           4

                                          3.5                           loop

                                           3
                                                    Level 1




                                    HQM
                                          2.5
                                                                                               h-dot loop
                                           2
                                                                            loop
                                          1.5

                                           1
                                                                                      loop
                                          0.5

                                           0
                                                0       2     4   6    8     10     12    14     16    18   20
                                                                      frequency rad/sec


Laboratory tracking tasks                 UH-60 hover task – ATTC/ATTH
                                                      SCAS
                    California Innovation Center

• The California Innovation Center provides a mechanism where industry
  and universities (UCD & CSU Sacramento) will come together to support
  the existing technology-focused missions at Beale Air Force Base. These
  collaborative efforts will support additional emerging technologies that will
  influence and embrace the future growth of autonomous and cyber systems.
                                    California Innovation Center

                                                            Safety & Reliability Issues
        Navigatio
           n
                                       Aircraft
                                    Airworthiness



FAA Airspace
Classification
      s
                                                                                                               FAR 91.113b
                                                                       Collision Avoidance with
                                                                 cooperative & non-cooperative aircraft   When weather conditions
                                                                                                          permit, regardless of whether an
                        Weather
                                                                                                          operation is conducted under
                                                    Compliance with
                       Avoidance?
                                                     14 CFR 91.113                                        instrument flight rules or visual
                                                                              Interoperability with
                                                                              manned / unmanned           flight rules, vigilance shall be
                                                                                     aircraft
                                        ATC                                                               maintained by each person
                                    Communications
                                                                                                          operating an aircraft so as to see
   Command & Control                                                                                      and avoid other aircraft.
        Link
                                             Operator                      Take-off & Landing
                                           Qualifications

								
To top