A workload adaptive haptic shared control scheme for semi-autonomous driving

•We develop a shared control scheme that adapts to the human operator's workload in real time.•We propose a heuristic design of the adaptive scheme based on human factors theories.•We estimate the human operators’ workload by using HMM to analyze their gaze trajectory.•The haptic control scheme results in significantly lower workload, higher trust in autonomy, better driving task performance and smaller control effort.