Exoskeleton robot and detection method for detecting human motion intention

Abstract

The invention discloses an exoskeleton robot which comprises an exoskeleton robot body, a sensing system, a control system and a drive system. The exoskeleton robot body comprises an upper part support structure, a hip rod piece, thigh rod pieces, shank rod pieces, foot components, a hip joint, knee joints and ankle joints; the sensing system comprises a thin film strain gage sensor for detectingthe sole pressure, an encoder for detecting a joint angle, a force sensor for detecting a force / moment of the joint part, an attitude sensor for detecting the kinematic velocity / the accelerated speed,a capacitance-type sensor for detecting the muscle tension, and a connecting buckle fastener; the capacitance-type sensor non-contacting with a human skin is fixed onto the connecting buckle fastenerand comprises at least one electrode patch; and the control system comprises a full-communication network computing module formed based on a CNN machine learning algorithm, is connected with the sensing system and can receive data acquired by the sensing system, and is connected with the drive system capable of driving the exoskeleton robot body to move.

Description

technical field [0001] The invention relates to the technical field of exoskeleton robots, in particular to an exoskeleton robot and a detection method for detecting human motion intentions. Background technique [0002] After decades of development, the research in the field of exoskeleton robot technology has made great progress, and some important results have been obtained, which provide a reference for the industrial application of exoskeleton. Many current exoskeleton systems, such as wearable exoskeleton robots, have achieved simple functional actions, but they are far from meeting the needs of the human body. [0003] On the one hand, due to the strong autonomy, high degree of freedom, complex information, and various actions of human body movement, in order to achieve good identification and prediction, it is necessary to have a sensory perception system that can reflect the system state to a greater extent, and the identification delay The time is small, the predi...

AI Technical Summary

Problems solved by technology

However, the effect of its motion delay and detection time is still poor

For another example, the AIDER exoskeleton robot of the University of Electronic Science and Technology of China is based on the intention recognition of the online SVM (Support Vector Machine), and the sample data can be updated every 20ms. The sample data consists of inertial measurement unit (IMU) and pressure shoe data. , the judgment time is ≥0.02s, and the effect is not very good; as another example, the Japanese HAL exoskeleton uses biosensors that can detect EEG signals and EMG signals to obtain the wearer’s movement intentions. Environmental changes are very sensitive, so the collection is difficult and the collection quality is not high, which limits the ability of the HAL exoskeleton to work in complex environments to a certain extent

Images