Motion intensity modeling and trajectory control of upper limb rehabilitation exoskeleton robot based on multi-modal information

The motion intensity of patient is significant for the trajectory control of exoskeleton robot during rehabilitation, as it may have important influence on training effect and human–robot interaction. To design rehabilitation training task according to situation of patients, a novel control method of rehabilitation exoskeleton robot is designed based on motion intensity perception model. The motion signal of robot and the heart rate signal of patient are collected and fused into multi-modal information as the input layer vector of deep learning framework, which is used for the human–robot interaction model of control system. A 6-degree of freedom (DOF) upper limb rehabilitation exoskeleton robot is designed previously to implement the test. The parameters of the model are iteratively optimized by grouping the experimental data, and identification effect of the model is analyzed and compared. The average recognition accuracy of the proposed model can reach up to 99.0% in the training data set and 95.7% in the test data set, respectively. The experimental results show that the proposed motion intensity perception model based on deep neural network (DNN) and the trajectory control method can improve the performance of human–robot interaction, and it is possible to further improve the effect of rehabilitation training.

PERCEIVED CHROMA AND HUE CHANGES OF COLOURS AT HIGH ILLUMINANCE LEVELS DUE TO HUNT EFFECT

The purpose of this study is to quantify the Hunt Effect in a range from indoor lighting levels to outdoor daylight levels so that a perception model of Hunt Effect for lighting can be developed with outdoor daylight as the reference. Our previous study experimentally quantified the perceived chroma changes due to the Hunt Effect at 100 lx and 1000 lx. To extend this to light levels closer to outdoor daylight, a vision experiment was conducted at ≈1000 lx and ≈6000 lx for red, green, yellow, and blue patches. A reference patch on one side of a double booth at 1000 lx was compared to a set of 20 test patches on the other side of the booth at ≈6000 lx using haploscopic view condition. Results showed that the perceived chroma changes are much smaller and insignificant compared to the results between 100 lx and 1000 lx found in our previous study.

Modeling Analysis and 3D Force Prediction of a Novel Piezoelectric Tactile Sensor

To measure three-dimensional (3D) forces efficiently and improve the sensitivity of tactile sensors, a novel piezoelectric tactile sensor with a “sandwich” structure is proposed in this paper. An array of circular truncated cone-shaped sensitive units made of polyvinylidene fluoride (PVDF) is sandwiched between two flexible substrates of polydimethylsiloxane (PDMS). Based on the piezoelectric properties of the PVD F sensitive units, finite element modelling and analysis are carried out for the sensor. The relation between the force and the voltage of the sensitive unit is obtained, and a tactile perception model is established. The model can distinguish the sliding direction and identify the material of the slider loaded on the sensor. A backpropagation neural network (BPNN) algorithm is built to predict the 3D forces applied on the tactile sensor model, and the 3D forces are decoupled from the voltages of the sensitive units. The BPNN is further optimized by a genetic algorithm (GA) to improve the accuracy of the 3D force prediction, and fairly good prediction results are obtained. The experimental results show that the novel tactile sensor model can effectively predict the 3D forces, and the BPNN model optimized by the GA can predict the 3D forces with much higher precision, which also improves the intelligence of the sensor. All the prediction results indicate that the BPNN algorithm has very efficient performance in 3D force prediction for the piezoelectric tactile sensor.

Applied Research on the Combination of Weighted Network and Supervised Learning in Acupoints Compatibility

To enhance the depth of excavation and promote the intelligence of acupoint compatibility, a method of constructing weighted network, which combines the attributes of acupoints and supervised learning, is proposed for link prediction. Medical cases of cervical spondylosis with acupuncture treatment are standardized, and a weighted network is constructed according to acupoint attributes. Multiple similarity features are extracted from the network and input into a supervised learning model for prediction. And, the performance of the algorithm is evaluated through evaluation indicators. The experiment finally screened 67 eligible medical cases, and the network model involved 141 acupoint nodes with 1048 edge. Except for the Preferential Attachment similarity index and the Decision Tree model, all other similarity indexes performed well in the model, among which the combination of PI index and Multilayer Perception model had the best prediction effect with an AUC value of 0.9351, confirming the feasibility of weighted networks combined with supervised learning for link prediction, also as a strong support for clinical point selection.