A -- ENGINEERING SUPPORT FOR THE INTELLIGENT FLIGHT CONTROL SYSTEMS (IFCS) F-15 837 PROGRAM

Notice Date

December 4, 2001

Contracting Office

NASA/Dryden Flight Research Center, Code A, P.O. Box 273, Edwards, CA 93523-0273

ZIP Code

93523-0273

Response Due

December 14, 2001

Point of Contact

Teresa M Hass, Contracting Officer, Phone (661) 276-2164, Fax (661) 276-2291, Email teresa.hass@dfrc.nasa.gov -- Roberta J. Ross, Contract Specialist, Phone (661) 276-3143, Fax (661) 276-2904, Email roberta.ross@dfrc.nasa.gov

E-Mail Address

Teresa M Hass (teresa.hass@dfrc.nasa.gov)

Description

NASA/DFRC plans to issue a Request for Quotation (RFQ) for the following Statement of Work for the IFCS (Intelligent Flight Control Systems) Program: Purpose: 7 The purpose of the IFCS program is to increase the ability of the control system to adapt to unknown failures and damage. This reconfigurable research topic increases the chances to land safely in the presence of a failure. The purpose of this SOW is to design the F-15 IFCS aircraft with a Neural Network control system with linear plant models and test the design in a nonlinear 6 DOF simulator. The tasks are as follows: Task 1) 7 The contractor shall review the Boeing/Dryden Control Law (CLAW) design for the F-15 IFCS. After the review the contractor shall give recommendations on the said design, and give strengths and weaknesses of the design. Task 2) 7 The contractor with his/her own Matlab/ Simulink (release 6.0 or 6.1) license will install Dryden's F-15 IFCS simulation (linear plant models, therefore no ITAR issues). Task 3) 7 Develop, install and optimize the Neural Network (NN) bases direct adaptive control algorithms. 7 3.1 Develop the control architecture in collaboration with Dryden, taking into account deviations from the pure model inversion that currently exist in the F-15 IFCS. Dryden will be the responsible party in the final design effort. 7 3.2 Modify where needed, the approach to the NN based adaptive control implementation to account for these deviations. The contractor will make the modifications when the NN control errors become unacceptable by Dryden's control engineers. 7 3.3 Incorporate NNs and control architecture in all three axes (pitch, roll and yaw). 7 3.4 Tune design parameters, including desired response characteristics, adaptation gains, activation potentials, and network architecture. 7 3.5 Develop preliminary performance evaluation for a spectrum of failure cases and flight conditions to be developed in concert with Dryden. With Dryden being the responsible party. Task 4) 7 Incorporate and design Pseudo Control Hedging (PCH) logic to robustify the design with respect to actuator nonlinearities (position and rate limits). 7 4.1 Develop the actuator models to be hedged. 7 4.2 Modify the current approach, as required, to account for aspects of the existing F-15 control law architecture that deviate from a pure model inversion based control architecture. 7 4.3 Develop a model for f_hat, the control map in theF-15 IFCS. This is required to compute the hedging signal. 7 4.4 Perform preliminary performance evaluation for a spectrum of failure cases and flight conditions that Dryden will be testing. Both, contractor and Dryden will develop, modify and test failure modes. Task 5) 7 Deliver preliminary NN design to Dryden, including reporting and oral briefings at Dryden. Task 6) 7 Support Piloted simulation evaluations. This support includes suggestions on the NN performance and improvement comments. Task 7) 7 On-site support during flight-test at Dryden. Includes post flight performance evaluation of the NN and control system; and real time NN performance tool development. Task 8) 7 Present lecture series on the F-15 IFCS NN design and lessons learned in the development of the controller. Provide to the government all the Matlab/Simulink tools that were required to develop the NN controller. Also have a simple Matlab/Simulink example as a design case, with and without control hedging. This procurement is being conducted under the Simplified Acquisition Procedures (SAP). NASA/DFRC intends to contract with Dr. Anthony Calise of 6980 Glenridge Drive, Atlanta, GA under the authority 10 USC 2304(c)(1), "Only One Responsible Source and No Other Supplies or Services will Satisfy Agency Requirements. FAR Part 6.302-1 (a)(iii) states that services may be deemed to be available only from the original source in the case of follow-on contracts for the continued provision of highly specialized services when it is likely that award to any other source would result in substantial duplication of cost to the Government that is not expected to be recovered through competition Dr. Calise has developed the neural network software used to demonstrate recovery from in-flight failures in the Air Force sponsored program called RESTORE. Dr. Calise is the predominate expert in the field of Neural Networks reconfigurable controls work. A neural network method has been tested on un-piloted vehicles such as the X-36 program. The next step in this research effort is to bring this technology to a piloted vehicle, where the certification level or reliability is much more difficult to obtain. Dr. Calise was instrumental in the design and testing of the un-piloted version of the neural networks and has developed special control design tools used to generate the flight control system. He has developed and designed Pseudo Control Hedging logic to robustify the design with respect to actuator nonlinearities (position and rate limits). Replacing Dr. Calise at this point will also require a considerable amount of redesign (generating computer codes and tools used with neural networks) along with additional simulation testing, all at NASA's expense. The dollar amount for a new neural network designer and the man-hours required for system redesigns, new software codes, and replicated simulation tests would be extensive. The cost of redesigning and building up the tools would result in substantial duplication of cost to the Government. The Government does not intend to acquire a commercial item using FAR Part 12. See Note 26. Interested organizations may submit their capabilities and qualifications to perform the effort in writing to the identified point of contact not later than 4:00 p.m. local time on December 14, 2001. Such capabilities/qualifications will be evaluated solely for the purpose of determining whether or not to conduct this procurement on a competitive basis. A determination by the Government not to compete this proposed effort on a full and open competition basis, based upon responses to this notice is solely within the discretion of the government. Oral communications are not acceptable in response to this notice. All responsible sources may submit an offer which shall be considered by the agency. An Ombudsman has been appointed. See NASA Specific Note "B". Any referenced notes can be viewed at the following URL: http://genesis.gsfc.nasa.gov/nasanote.html

Web Link

Click here for the latest information about this notice (http://nais.msfc.nasa.gov/cgi-bin/EPS/bizops.cgi?gr=D&pin=24#99530)

Record

Loren Data Corp. 20011206/ASOL005.HTM (D-338 SN5143Y4)