Research And Development Of The Seat Pressure Distribution Measurement System Used For Driver Fatigue Detection Research

In recent years,driver fatigue detection has become a hot research field of vehicle active safety technology.It will start a new research direction to detect driver fatigue based on seat pressure distribution.When driving on the road,driving activities,like manipulating the steering wheel,pressing the clutch pedal,the brake pedal and the gas pedal,will change drivers’ posture and the posture changes will be embodied on the changes of pressure on the seat.When drivers get into fatigue,the driving operations will become abnormal and the pressure distribution will change correspondingly.Hence,the changes of the seat pressure distribution can be used to recognize drivers’ driving operations,in order to find the fatigue features and realize driver fatigue detection.A seat pressure distribution measurement system used for driver fatigue detection research is designed in this paper.This system is important for both driver fatigue detection research and application.The main work and results are as follows:(1)The integral system structure has been designed,including the design and debug of the hardware and the software.After Tekscan pressure mapping sensor is chosen,the data acquisition hardware board is designed reasonably,which consists main control module,analog switch module,communication module and power module,etc.And efficient embedded control program is designed.The seat pressure data processing software is designed by VB and it realizes the key functions of graphic user interface,data visualization and data storage,etc.The hardware and software debugging is completed by acquiring pressure data on the driving simulation platform,and the system can finally operate normally which validates the rationality and feasibility of the system design.(2)Recognizing driving operations based on seat pressure distribution has been exploringly researched.This paper classifies driving operations into single-operation mode and multi-operation mode.Recognition of single driving operation mode based on seat pressure distribution is researched.A dynamic coordinate model is established and 27 feature parameters are extracted from the model.After the contrast experiments of the four pattern recognition algorithms on 1173 samples,k-nearest neighbor classifier is found to be the best recognition method.