Research On The Detection Of Driver’s Hand Off Steering Wheel Based On Data Fusion Of Smartwatch And Smartphone

Whether the driver’s hand is off the steering wheel is very important to ensure driving safety,detecting whether the driver’s hand is off the steering wheel in time and warning the driver can improve the driver’s self-alertness and the driving skill.However,due to the impact of the vehicle’s movement and the significant variation across different driving environment,detecting the position of the driver’s hand is challenging.Existing computer vision-based method is easily affected by environmental factors such as illumination and touch-sensing device-based method requires additional specific sensors such as pressure sensors on the steering wheel,resulting in additional costs.Therefore,it is of great significance to study a method for detecting whether the driver’s hand is on/off the steering wheel that does not rely on computer vision and additional sensors for monitoring driver unsafe behavior and preventing traffic accidents.Considering that smartwatch and smartphone are now widespread and can effectively monitor the movement of targets,this paper uses smartwatch and smartphone to collect drivers’ hand behavior and vehicle state data respectively.A more convenient and practical method of detecting whether the driver’s hand is on/off the steering wheel is established by detecting whether the driver’s hand is off the steering wheel when the hand is still or moving.The main research work of this paper includes the following three aspects:Firstly,considering that whether the driver’s hand is on/off the steering wheel when the hand is still will reflect different data characteristics on the sensor,this paper extracts a variety of feature indexes from the sensor data based on the driver’s forearm posture and vehicle vibration signals.Then,Based on the anomaly detection algorithm,a detection model for the driver’s hand off the steering wheel is established when the hand is still.Secondly,In order to eliminate the impact of the phone’s attitude on the data,this paper implements the conversion of the data collected by the smartphone from the device coordinate system to the world coordinate system based on the API of Android system,and extracts features based on the vehicle physical model in the world coordinate system.Based on the vehicle steering recognition model,a detection model for the driver’s hand off the steering wheel is established when the hand is moving.Thirdly,in order to establish a detection model with high accuracy,this paper implements driver hand movement detection based on the absolute acceleration difference between the smartwatch and smartphone,and divides the driver’s hand state into a relatively stationary state and a motion state to complete the detection method of driver’s hand off steering wheel.On this basis,the automatic update of the training data set is further realized,and the method is improved.