This paper makes two scientific contributions to the field of exoskeleton-based action and movement recognition. First, it presents a novel machine learning pattern recognition-based framework that can detect wide range actions movements - walking, walking upstairs, downstairs, sitting, standing, lying, stand sit, sit stand, lie, lie with an overall accuracy 82.63%. Second, comprehensive compar...