This paper proposes a specific method of balance control learning at a half-rising posture using an assistive system for improving the pulling chair movement, and reports a case study of a quadriplegic wheelchair user. The system provides informatics-based biofeedback and user weight support. It does not require attaching any equipment or sensor to the user’s body. Although our participant was able to transfer from a wheelchair to a chair, she was not able to pull a chair until the desired position. We analyzed the physical kinematics and kinetics during pulling chair movement and developed a balance control learning method that enables the user to move the center of gravity (COG) backward when assuming a deep half-rising posture. We conducted an experiment on a quadriplegic participant once a week for four weeks. Before learning, the participant was not able to realize the pulling chair movement even though a lot of time was provided. However, after learning, the participant could move the COG backward when attempting to realize the half-rising posture, and pull a chair completely. This case proved our hypothesis that a half-rising posture with backward shifting of the center of pressure is effective in improving the pulling chair movement. Through the experiment, we achieved proof of concept of the proposed learning method and the first case study of successful application of the method.