Yiqing Zhu

Aiming at the problem that the pedestrian navigation system cannot effectively perform positioning and navigation when foot-mounted IMU exceeds its measurement ranges during vigorous activities or collisions with obstacles, a novel pedestrian navigation method is proposed based on construction of adapted virtual inertial measurement unit (VIMU) assisted by gait type classification. The method utilizes attention-based convolutional neural network (CNN) to classify gait types of the pedestrian equipped with the navigation system. Then, with the inertial data from pedestrian's thigh and foot collected synchronously via actual IMUs as training and testing samples, correspondent Resnet-Gated Recurrent Unit (Resnet-GRU) deep hybrid neural network models are built according to diverse gait types. Through these models, virtual foot-mounted IMU is constructed for positioning in case of physical foot-mounted IMU overrange. The experimental results show that, the proposed VIMU construction method can effectively improve the performance and reliability of zero velocity update-based pedestrian navigation system when the pedestrian moves or hits some obstacles with foot vigorously, thus enhancing the adaptability of the navigation system in complex and unknown terrains. The positioning error under comprehensive gait types is about 1.45% of the total walking distance, which meets the accuracy requirement in military and civilian applications.

When the wearer of Wearable Human-Machine Interaction System (WHMIS) is talking a strenuous walk, the Inertial Measurement Units (IMUs) installed on the extremities are likely to be beyond the range of design specification of the IMU sensors. This overrange leads to poor navigation performances of the system. To overcome the problem, this paper proposes a deep learning-based method to realize robust autonomous navigation. Specifically, the Faster Region-Convolutional Neural Network (R-CNN) is built to recognize the gait types of the wearer. For each gait type, the inertial data of the wearer's thigh and the foot are collected at the same frequency as the training samples, and then used to construct a correspondent Gated Recurrent Unit-Support Vector Regression (GRU-SVR) deep hybrid neural network model for virtual IMU simulation. Then, the virtual foot-mounted IMU will be constructed in real-time by the physical thigh-mounted IMU based on the neural network models mentioned above. Once the over-range of the physical foot-mounted IMU is detected, the virtual foot inertial information will be used to reconstruct the WHMIS navigation system. Experimental results show that the proposed navigation method has the ability of fault tolerance in complex gaits, and the 3D positioning performance of the reconstructed navigation system is approximately equivalent to that of the fault-free WHMIS navigation system.