SysIDLib: A High-Level Synthesis FPGA Library for Online System Identification

Model accuracy is the most important step towards efficient control design. Various system identification techniques exist which are used to estimate model parameters. However, these techniques have their merits and demerits which need to be considered before selecting a particular system identification technique. In this paper, various system identification techniques as the Kalman filter (EKF), recursive least square (RLS) and least mean square (LMS) filters are used to estimate the parameters of linear (DC motor) and nonlinear systems (inverted pendulum and adaptive polynomial models). FPGAs are widely used for rapid prototyping, real-time and high computationally demanding applications. Therefore, a real-time FPGA-in the loop architecture has been used for evaluating each identification algorithm of the SysIDLib library. The identification algorithms are evaluated regarding the convergence rate, accuracy and resource utilization performed on a system-onchip (SoC). The results have shown that the RLS algorithm estimated approximately the parameter values of a nonlinear system. However, it requires up to 17% less lookup-tables, 5.5% less flip-flops and 14% less DSPs compared to EKF with accurate results on the programmable logic (PL).

[1]  Alfonso Gómez-Espinosa,et al.  Nonlinear identification of inverted pendulum system using Volterra polynomials , 2016 .

[2]  Harish Kumar Sahoo,et al.  Harmonics and decaying DC estimation using Volterra LMS/F algorithm , 2018, 2016 IEEE 1st International Conference on Power Electronics, Intelligent Control and Energy Systems (ICPEICES).

[3]  Zoran A. Salcic,et al.  A floating-point FPGA-based self-tuning regulator , 2006, IEEE Transactions on Industrial Electronics.

[4]  Eric Monmasson,et al.  FPGA-Based Implementation of Dual Kalman Filter for PV MPPT Applications , 2017, IEEE Transactions on Industrial Informatics.

[5]  Roberto Roncella,et al.  Hardware-in-the-loop simulation of FPGA-based state estimators for electric vehicle batteries , 2016, 2016 IEEE 25th International Symposium on Industrial Electronics (ISIE).

[6]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .

[7]  Diana Göhringer,et al.  System Identification using LMS, RLS, EKF and Neural Network , 2019, 2019 IEEE International Conference of Vehicular Electronics and Safety (ICVES).

[8]  Rene de Jesus Romero-Troncoso,et al.  Special purpose processor for parameter identification of CNC second order servo systems on a low-cost FPGA platform , 2010 .

[9]  M. V. Vaidyan,et al.  An FPGA-based parallel architecture for on-line parameter estimation using the RLS identification algorithm , 2014, Microprocess. Microsystems.

[10]  Diana Göhringer,et al.  Dynamic tunable and reconfigurable hardware controller with EKF-based state reconstruction through FPGA-in the loop , 2018, 2018 International Conference on ReConFigurable Computing and FPGAs (ReConFig).

[11]  Yusuke Sato,et al.  Real-Time UAV Attitude Heading Reference System Using Extended Kalman Filter for Programmable SoC , 2017, 2017 IEEE 11th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC).

[12]  Luis Angel Barragan,et al.  High-Level Synthesis for Accelerating the FPGA Implementation of Computationally Demanding Control Algorithms for Power Converters , 2013, IEEE Transactions on Industrial Informatics.

[13]  Ajit Kumar Sahoo,et al.  Adaptive Volterra Modeling For Nonlinear Systems Based on LMS Variants , 2018, 2018 5th International Conference on Signal Processing and Integrated Networks (SPIN).

[14]  Jann N. Yang,et al.  An adaptive extended Kalman filter for structural damage identifications II: unknown inputs , 2007 .