In recent years, considerable research has been conducted to support human movement. However, most studies have been conducted on healthy subjects, and few have been conducted on subjects with disabilities. The purpose of this study is to propose a heart rate estimation model for wheelchair driving using 6-axis sensors to provide exercise support to wheelchair users. The heart rate was estimated using the machine learning from data from a 6-axis sensor attached to the seat of a wheelchair. Input data to the suggested model were the driving direction acceleration, slope of road and oxygen uptake. When the suggested model was applied to the heart rate estimation in driving wheelchair, it was confirmed to be possible of estimation with 9.21bpm Mean Absolute Error(MAE). Furthermore, the estimation accuracy could be improved by the advanced calculation of the oxygen uptake and introducing the parameters of the individual physical ability levels.