Clinical evaluation of frailty in the elderly is the first step to decide the degree of assistance they require. Advances in robotics make it possible to turn a standard assistance device into an augmented device that may enrich the existing tests with new sets of daily measured criteria. In this paper we use a standard 4 wheeled rollator, equipped with a Kinect and odometers, for biomechanical gait analysis. This paper focuses on the method we develop to measure and estimate legs and feet position during an assisted walk. The results are compared with motion capture data, as a ground truth. Preliminary results obtained on four healthy persons show that relevant data can be extracted for gait analysis. Some criteria are accurate with regards to the ground truth, eg. foot orientation and ankle angle.