Research On Motion Tracking And Human-computer Interaction System Based On Acoustic Signal

The research of smart device interaction system is always iterating with the progress of the times.In the process of transition from the button era to the touch screen era,the convenience of human-computer interaction has been greatly improved.However,the speed of development and popularization of mobile devices is unprecedented,and the emergence of various novel applications puts forward higher requirements for human-computer interaction.Currently,the interactive methods of mobile devices(such as mobile phones)are mainly to slide,tap and scroll the screen with your fingers.This kind of interaction is difficult to apply to emerging applications,such as video games,virtual reality(VR),augmented reality(AR),and has great limitations on non-contact control.Fortunately,with the rapid development of wireless sensing technology in recent years,moving tracking has been recognized as one of the most promising user interaction technologies to promote these applications.On this basis,users can interact with electronic devices without touching or adding additional devices.In this article,a collaborative gesture tracking method based on acoustic signals is proposed by using audio equipment that is ubiquitous and inexpensive in mobile devices to implement a human-computer interaction system CTrack(Collaborative Tracking)without additional equipment on a two-dimensional plane.CTrack can realize human-computer interaction based on one or both hands of the user on a two-dimensional plane.We have verified through experiments that we can achieve millimeter-level recognition ranging and positioning accuracy,and a recognition accuracy rate of more than 90%.This article first introduces the current research status of target tracking at home and abroad,and then introduces the principle and necessary basic knowledge of CTrack target tracking,and then introduces the design ideas and usage scenarios in detail.Finally,a system prototype analysis CTrack is implemented on a smart phone.System performance.The main work and contributions of this paper are as follows:1.Designed and implemented a collaborative target tracking system CTrack based on acoustic signals.The system converts the smart phone into an active sonar system.Specifically,it uses a speaker array to transmit a high-frequency chirp signal,and the microphone array receives echoes reflected by the target.Then,the position of the target is accurately located on the two-dimensional plane according to the geometric relationship formed by the speaker array,the microphone array and the target.In addition,according to the continuity of the target motion and the regional characteristics of the acoustic echo,the motion of the dual target is distinguished and tracked.We evaluated CTrack on Nexus 6P.The experimental results show that CTrack has good ranging,motion tracking,anti-environmental interference performance,and has good robustness and universality in a two-dimensional plane.Under the premise of limiting the user's operating area,the interactive performance based on dual goals also has the desired performance.2.Proposed Neighbor Frame Constraint(NFC)and Adaptive Localization Correction(ALC)algorithms to ensure positioning stability and reliability.The NFC algorithm avoids the error of the ranging result from interfering with the accuracy of the experiment.The ALC algorithm dynamically selects different audio equipment groups for positioning according to the error size calculated by the error NFC,thereby solving the large positioning error of the microphone due to the interference of the mobile phone shell.