Real-time Numerical Optimization in Reconfigurable Hardware with Application to Model-Predictive Control

Lead Research Organisation: Imperial College London
Department Name: Electrical and Electronic Engineering

Abstract

This proposal is concerned with the hardware acceleration of iterative numerical algorithms, with a focus on model predictive control implementations. Such model predictive controllers typically require the solution of a quadratic progamming problem every sample period. The solution of the quadratic programming problem typically requires several multidimensional Newton optimizations, each of which requires the solution of many systems of linear equations. Thus the lessons learned will be applicable to a wide class of numerical algorithms arising in practical problems within and beyond Control.The main adventurous feature of the approach from the digital electronics perspective is the potential to use Control and Systems theory to inform one of the central design problems in custom reconfigurable computing: efficient silicon utilization through appropriate finite precision number representation. In sequential (single core) computer architecture, questions of numerical precision have, by and large, been answered through the introduction of area costly high-precision IEEE compliant arithmetic units. In modern computing systems, whether FPGA-based or manycore, attention is now turning to how to make the most effective use of the silicon available for computation and, in this context, questions of numerical accuracy requirements are arising once more.The proposed approach forms a radical departure from standard industrial and academic practice in both model predictive control (MPC) and digital electronics. The main adventurous feature of the approach from the end-user perspective is the utilization of reconfigurable hardware devices, namely Field-Programmable Gate Arrays (FPGAs), to implement model predictive controllers operating at high sample rates, allowing MPC to be utilized in application areas where the computational load has been considered too great until now, such as spacecraft, aeroplanes, uninhabited autonomous vehicles, automobile control systems and gas turbines. From the theoretical perspective, the main adventure in Control is in the development of novel formulations that explcitly take advantage of parallel computational architectures.The development of a methodology to tackle this problem will involve highly novel research areas resulting from the application of control theoretic ideas to hardware development, as well as the application of hardware implementation methodologies to control system design. In particular, this proposal is the first to investigate massively parallel real-time numerical optimization on FPGAs, the first to apply control-theoretic techniques to determine appropriate number systems in custom hardware designs, and the first to study the tradeoff between circuit parallelism and numerical accuracy within a closed-loop behavioural context.As a result, this proposal directly falls within the scope of EPSRC's recently signposted Microelectronics Grand Challenge 3 - Moore for Less.

Publications

10 25 50

publication icon
Boland D (2011) Optimizing memory bandwidth use and performance for matrix-vector multiplication in iterative methods in ACM Transactions on Reconfigurable Technology and Systems

publication icon
Boland D (2011) Bounding Variable Values and Round-Off Effects Using Handelman Representations in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

publication icon
Buchstaller D (2012) Sampling and controlling faster than the computational delay in IET Control Theory & Applications

publication icon
Constantinides G (2011) Numerical Data Representations for FPGA-Based Scientific Computing in IEEE Design & Test of Computers

publication icon
Hartley E (2014) Predictive Control Using an FPGA With Application to Aircraft Control in IEEE Transactions on Control Systems Technology

publication icon
Hasan A (2013) Control-Theoretic Forward Error Analysis of Iterative Numerical Algorithms in IEEE Transactions on Automatic Control

 
Description That reconfigurable computing can allow optimization based control methods to be used in high performance settings.

That control theory can be used to inform design metrics in cyber physical systems.
Exploitation Route This work has spawned a number of ongoing special sessions in the major control conferences - several groups in academia and industry (e.g. ABB) now actively work taking this forward.
Sectors Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Electronics,Energy

 
Description Several companies have taken up this work (notably ABB) and one of the researchers on this project (Juan Jerez) has launched a startup in this area (embotech).
First Year Of Impact 2010
Sector Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Electronics,Energy
Impact Types Societal,Economic

 
Description FP7 STREP - EMBOCON
Amount £685,139 (GBP)
Funding ID EMBOCON 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 10/2010 
End 09/2013
 
Description Siemens Research Award
Amount £174,767 (GBP)
Organisation Siemens AG 
Sector Private
Country Germany
Start 10/2013 
End 09/2016
 
Description Agility - MPC 
Organisation Agility Design Solutions
Country United States 
Sector Private 
PI Contribution Research Output
Collaborator Contribution Staff time
Impact Agility DS was bought out
Start Year 2009
 
Description ESA - MPC 
Organisation European Space Agency
Country France 
Sector Public 
PI Contribution Research output
Collaborator Contribution Staff time
Impact Research output, follow-on funding
Start Year 2009
 
Description Mathworks - MPC 
Organisation The Mathworks Ltd
Country United Kingdom 
Sector Private 
PI Contribution Research output
Collaborator Contribution Staff time, licences
Impact Publications
Start Year 2009
 
Description NTU - MPC 
Organisation Nanyang Technological University
Country Singapore 
Sector Academic/University 
PI Contribution Research output
Collaborator Contribution Staff time
Impact Joint publications, student exchanges
Start Year 2009
 
Description Xilinx - MPC 
Organisation Xilinx Corp
Country United States 
Sector Private 
PI Contribution Research output
Collaborator Contribution Staff time, licences, devices
Impact Publications
Start Year 2009