Research On The Evaluation And Prediction Of The Human’s Workload Capacities For The In-orbit Manipulation

The objective and quantitative researches of human’s workload capabilities during spaceflight missions should be strengthen in order to facilitate the exploration and utilization of space resources and promote the development of space technology. This dissertation carried out research with two models: the evaluation model and the prediction model.Firstly, the evaluation model of human’s workload capabilities for manned space flight missions was established. The basic characteristics of this evaluation model should have noninterference, nonintervention, invasion, quantification, objectivity and accuracy. This dissertation first determined the structure of the model which was based on surface Electromyography(s EMG) and muscle force. Then under the guidance of multiple features theory, the non-stationary and non-linearity of s EMG were depicted by the time domain, frequency domain and bi-frequency domain analyses. The regularities of variation of amplitude, frequency and phase parameters with changes of workload capabilities were investigated. Using the statistical analyses of muscle force, this dissertation presented the mathematical descriptions of physical attributes about the human’s workload capabilities. Then it explored the relationships between different attributes for muscle force data under different operation conditions. Furthermore, it compared the differences among operation capabilities under different operation conditions, and analyzed the impact factors that influenced the changes of the workload capabilities. After the features of the s EMG signal and muscle force were extracted, the root mean square(RMS) and integrated EMG(i EMG) of time domain, the mean power frequency(MPF) and median frequency(MF) of frequency domain, as well as the diagonal slice spectrum and axial slice spectrum of bispectra were combined with support vector machine(SVM) to identify and classify movement patterns for the future in-orbit manipulation. This method has higher accuracy for the recognition rate. In addition, a mathematical model of biceps brachii based on musculoskeletal physiology and anatomy was established to demonstrate which the parameters of amplitude, frequency and phase regularly vary with changes of workload capabilities.Secondly, the prediction model of human’s workload capabilities for the in-orbit manipulation was studied. The near future development tendency of human’s workload capabilities was predicted by Elman neural net. The MPF and MF results of biceps brachii’s s EMG from single left and single right upper limb, and corresponding muscle force data were selected as the input parameters of Elman model to forecast the workload capabilities. The prediction model includes four layers structure, there are: the input layer, hidden layer, connectivity layer and output layer. This model solved the prediction problem of workload capabilities which have multiple variables, tight coupling, non-linearity and dynamic.Finally, the experimental method for human’s workload capabilities for the in-orbit manipulation under the earth’s gravity and simulated weightlessness environments was explored. Experiments involved five-level operation conditions, three-level position restrictions, two-level operation postures, and four-level operation directions. During the experiments, the three-dimensional muscle force and the four-channel s EMG signal were collected. The evaluation and prediction models were validated by the test data.Motivated by practical requirement of growing depth and breadth of space exploration in China, this dissertation focuses on the issues of the evaluation and prediction for the workload capacities of human during space manipulation. This dissertation has preliminary built the methods of real time monitoring, comprehensive evaluation, effective classification and identification, as well as the trend prediction. The achievements will provide technical support for the establishment and improvement of man-machine functional allocation in the astronaut—spacecraft system.