@Kai
So, you have experience in a similar project. One question even if we use voltage directly read from load and from series resistor, it has to be read by some sort of diff amp. Those diff amps introduce some error in phase relative to real voltage waveform that the load senses and also current waveform. We thought to use the same arrangement and type of diff amp both for reading voltage and current with the hope that these errors cancel each other. Did you done the same? And a new problem has been raised by one of my colleagues to keep us worry. What if the load itself has some significant capacitance for example 100nF, then the stability problem will become even more complex…Oh!God why I chose this job!!! I could be a happy shopkeeper! ;-)