Walking Simulation for VR Game Character Using Remote Sensoring Device Based on AHRS-Motion Recognition

Autor: J. Y. Kim, W. H. Lee
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Zdroj: IEEE Access, Vol 7, Pp 19423-19434 (2019)
Druh dokumentu: article
ISSN: 2169-3536
DOI: 10.1109/ACCESS.2018.2878237
Popis: In this paper, the remote sensoring device for controlling the avatar in the virtual reality (VR) game which is being worn by the head-mounted display (HMD) was investigated. This remote sensoring device, which people use while sitting on a swivel chair, minimizes sensor attachment, and removes parameters for pelvic rotation angle in standing position. The rotating chair, which also functions as the center-weighted stationary belt of the conventional treadmill-type input device, has the advantage that the head trekking function mounted on the HMD can be utilized freely and the balance feeling of the user can be stabilized in a sitting position on the chair. In this posture, the sensoring input device was developed using Bluetooth wireless communication to freely turn and walk. Using the data from the inertial measurement device, which was a combination of the speedometer and the accelerometer, a sensoring input device for recognizing the user's leg motion was created. The running speed of the recognized leg was designed to be close to the average walking speed and the running speed of the person, so that the running and walking in the game can be separated and implemented. The motion velocity of the leg would be calculated through each speedometer measurement, the value would be integrated, and the error would be corrected by the Kalman filter to obtain the user's knee angle. In other words, applying a Kalman filter to data obtained from sensors, such as accelerometers, angular speedometers, and geomagnetic sensor can determine the direction of motion of the legs. This integrated system is called the attitude and heading references system. The knee joint angle and the value of the pressure sensor were used as threshold values to determine the state of the leg when walking, and the angular velocity derived from the movement of the leg was used to determine the moving speed of the avatar in the VR game. The recognition rate was calculated by comparing the number of times the user attempts to move the avatar by moving the avatar and the number of times the avatar of the game was moved, and the distance of the avatar was measured to evaluate the performance of the proposed system.
Databáze: Directory of Open Access Journals