Bio-Inspired Real-Time Prediction of Human Locomotion for Exoskeletal Robot Control

Abstract

Abstract:Human motion detection is of fundamental importance for control of human-robot coupled systems such as exoskeletal robots. Inertial measurement units have been widely used for this purpose, although delay is a major challenge for inertial measurement unit-based motion capture systems. In this paper, we use previous and current inertial measurement unit readings to predict human locomotion based on their kinematic properties. Human locomotion is a synergetic process of the musculoskeletal system characterized by smoothness, high nonlinearity, and quasi-periodicity. Takens' reconstruction method can well characterize quasi-periodicity and nonlinear systems. With Takens' reconstruction framework, we developed improving methods, including Gaussian coefficient weighting and offset correction (which is based on the smoothness of human locomotion), Kalman fusion with complementary joint data prediction and united source of historical embedding generation (which is synergy-inspired), and Kalman fusion with the Newton-based method with a velocity and acceleration high-gain observer (also based on smoothness). After thorough analysis of the parameters and the effect of these improving techniques, we propose a novel prediction method that possesses the combined advantages of parameter robustness, high accuracy, trajectory smoothness, zero dead time, and adaptability to irregularities. The proposed methods were tested and validated by experiments, and the real-time applicability in a human locomotion capture system was also validated.

Wearable Sensor-Based Real-Time Gait Detection: A Systematic Review

“…Recent advancements in technology have led to reliable, affordable, and wearable sensors for gait analysis that enable its use outside of a laboratory environment and during activities of daily living. One of its primary uses has been in diagnosing walking impairment in people with gait disabilities [ 1 , 2 , 3 , 4 ] and inspires control mechanisms of exoskeletons [ 5 , 6 ] and prostheses [ 7 ], among other applications [ 8 , 9 , 10 , 11 , 12 ]. More specifically, real-time gait analysis has proven essential in applications necessitating real-time control such as exoskeletons and prostheses, as well as gait rehabilitation involving Functional Electrical Stimulation (FES) [ 13 , 14 ] or Epidural Electrical Stimulation (EES) [ 15 ].…”

Exoskeleton Follow-Up Control Based on Parameter Optimization of Predictive Algorithm

“…Although the Takens algorithm cannot predict that all motion sensor data are completely correct, some papers also give an optimization method based on this algorithm [ 14 ], which makes the Takens algorithm used in embedded systems become possible. Whitney's topological embedding theorem shows that the phase space of the original system can be reconstructed by any value of the delay time of the single variable time series in the case of noiseless and infinitely long time series.…”

“…Duan et al have found that inappropriate historical vector dimension and embedding dimension will cause irrelevant data to enter the reconstruction delay vector, resulting in lower prediction accuracy [ 14 ]. While inappropriate prediction duration and optimal matching number will cause data jitter and affect the smoothness of the predicted data.…”

“…The results showed that Takens is more stable, accurate, and suitable than LMS, but it did not solve the zero dead time problem of the prediction algorithm. So, Duan et al [ 14 ] use the combination of the Takens prediction algorithm and Newton-based-method to solve the problem of the zero dead time of the prediction algorithm but did not make progress in solving the parameter tuning process.…”

Special Feature on Bio-Inspired Robotics

“…Takens' reconstruction method is used to characterize quasi-periodicity and nonlinear systems. With Takens' reconstruction framework, they develop following methods, including Gaussian coefficient weighting and offset correction which is based on the smoothness of human locomotion, Kalman fusion with complementary joint data prediction and united source of historical embedding generation which is synergy-inspired, and Kalman fusion with the Newton-based method with a velocity and acceleration high-gain observer also based on smoothness [29]. Jiang et al present a hardware-based method that utilizes a shielded drive circuit to eliminate extraneous interferences on biopotential signal recordings, while also preserving all useful components of the target signal.…”