As was previously said, the procedure does not exist in R as of now. I myself am an avid R user and needed to use Westerlund's error-correction panel cointegration test for my research. I resorted to using Stata for that and stuck with R for everything else. I actually started working on a Westerlund function in R, but left the project some time ago due to other things I was working on. If anyone would like to collaborate on that, please contact me.