Computing Exp (A) and Its Integral

Research output: Book/ReportReport


This report describes one algorithm to compute exp(A) and one algorithm to compute both exp(A) and its integral. The method used in the two algorithms is finite series approximation, where the matrix is scaled before the expansion. The choice of the number of terms in the series expansion is discussed in detail, and results of numerical investigations performed on the PDP 15/35 computer are presented. The conclusion is that nine terms is a suitable choice on a computer with a floating point accuracy of 7-8 decimal digits.


  • Claes Källström
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Control Engineering
Original languageEnglish
PublisherDepartment of Automatic Control, Lund Institute of Technology (LTH)
Publication statusPublished - 1973
Publication categoryResearch

Publication series

NameResearch Reports TFRT-3053
ISSN (Print)0346-5500

Total downloads

No data available