Hand Posture Monitoring And Reconstruction For Forestry Mechanical Gesture Control

Forestry workers can remotely control intelligent forestry machines with different gestures to perform various types of operations,which can not only eliminate the potential safety hazards caused by the unstable machine performance of traditional forestry machines to forestry workers,but also greatly improve the working environment of forestry workers,keep them away from rugged forest shrubs and long-term noise environments.In order to realize the gesture control of the intelligent forestry machinery,it is necessary to use the computer to recognize the hand movement posture of the forestry workers in time,and then the computer controls the intelligent forestry machinery to perform related operations in real time according to the hand posture of the forestry workers and the preset programs.Based on optical fiber curvature sensing technology,this paper proposes a data glove for monitoring and reconstructing hand posture for forestry machinery gesture control.Monitor and simulate the hand posture of forestry workers,laying a foundation for further realizing the control of forestry machinery with gestures.The main research contents of this paper are as follows:(1)The characteristics and advantages of the fiber optic sensor are analyzed,and a fiber optic sensor tape is fabricated based on the fiber optic curvature sensor with a sensitizing structure,which provides a basis for the subsequent production of data gloves.A reference optical signal is introduced for the optical fiber sensing signal in the optical fiber sensor tape to improve the anti-interference ability of the optical fiber sensing signal.At the same time,the assembly scheme of the optical fiber sensor and the packaging technology of the optical fiber sensor tape are studied.(2)By analyzing the human hand structure and the correlation between the finger joints,the arrangement scheme of measuring the bending state of the finger by using the optical fiber sensor tape is studied.The roll,pitch and yaw angles of the wrist joint of the hand are measured by the orientation sensor.In this paper,a signal processing circuit and a signal processing program are designed and produced,and a communication transmission system for data signals is established.The information collected by the monitoring system is transmitted to the computer to provide basic data for the reconstruction of the hand posture.(3)The calibration test is carried out on the bending angle of the index finger joint and the pitch angle of the wrist joint.The functional relationship between the system output value and the actual bending angle of the finger and the pitch angle of the wrist joint is obtained,and the fitting curve is drawn.The fitting curve is used to repeat the monitoring test of the bending angle of the finger joint and the pitch angle of the wrist joint,and the monitoring accuracy of the system is analyzed.The experimental results show that the maximum error of monitoring the hand posture using the optical fiber sensor and the direction sensor is π/40 and π/120,respectively.(4)Real-time reconstruction of hand pose.Use 3ds Max software to make a virtual hand model and perform bone binding and skinning processing on the model.Import the prepared hand model into the Unity software.In the Unity software,each main motion joint of the hand is associated with the monitoring data of the corresponding finger of the data glove to realize the real-time simulation reconstruction of the hand posture.