The vast majority of products and processes in industry and academia require human interaction. Thus, digital human models (DHMs) are becoming critical for improved designs, injury prevention, and a better understanding of human behavior. Although many capabilities in the DHM field continue to mature, there are still many opportunities for improvement, especially with respect to posture- and motion-prediction. Thus, this thesis investigates the use of artificial neural network (ANN) for improving predictive capabilities and for better understanding how and why human behave the way they do. With respect to motion prediction, one of the most challenging opportunities for improvement concerns computation speed. Especially, when considering dynamic motion prediction, the underlying optimization problems can be large and computationally complex. Even though the current optimization-based tools for predicting human posture are relatively fast and accurate and thus do not require as much improvement, posture prediction in general is a more tractable problem than motion prediction and can provide a test bead that can shed light on potential issues with motion prediction. Thus, we investigate the use of ANN with posture prediction in order to discover potential issues. In addition, directly using ANN with posture prediction provides a preliminary step towards using ANN to predict the most appropriate combination of performance measures (PMs) - what drives human behavior. The PMs, which are the cost functions that are minimized in the posture prediction problem, are typically selected manually depending on the task. This is perhaps the most significant impediment when using posture prediction. How does the user know which PMs should be used? Neural networks provide tools for solving this problem. This thesis hypothesizes that the ANN can be trained to predict human motion quickly and accurately, to predict human posture (while considering external forces), and to determine the most appropriate combination of PM(s) for posture prediction. Such capabilities will in turn provide a new tool for studying human behavior. Based on initial experimentation, the general regression neural network (GRNN) was found to be the most effective type of ANN for DHM applications. A semi-automated methodology was developed to ease network construction, training and testing processes, and network parameters. This in turn facilitates use with DHM applications. With regards to motion prediction, use of ANN was successful. The results showed that the calculation time was reduced from 1 to 40 minutes, to a fraction of a second without reducing accuracy. With regards to posture prediction, ANN was again found to be effective. However, potential issues with certain motion-prediction tasks were discovered and shed light on necessary future development with ANNs. Finally, a decision engine was developed using GRNN for automatically selecting four human PMs, and was shown to be very effective. In order to train this new approach, a novel optimization formulation was used to extract PM weights from pre-existing motion-capture data. Eventually, this work will lead to automatically and realistically driving predictive DHMs in a general virtual environment.

Artificial neural networks (ANNs) have been used successfully in various practical problems. Though extensive improvements on different types of ANNs have been made to improve their performance, each ANN design still experiences its own limitations. The existing digital human models are mature enough to provide accurate and useful results for different tasks and scenarios under various conditions. There is, however, a critical need for these models to run in real time, especially those with large-scale problems like motion prediction which can be computationally demanding. For even small changes to the task conditions, the motion simulation needs to run for a relatively long time (minutes to tens of minutes). Thus, there can be a limited number of training cases due to the computational time and cost associated with collecting training data. In addition, the motion problem is relatively large with respect to the number of outputs, where there are hundreds of outputs (between 500-700 outputs) to predict for a single problem. Therefore, the aforementioned necessities in motion problems lead to the use of tools like the ANN in this work. This work introduces new algorithms for the design of the radial-basis network (RBN) for problems with minimal available training data. The new RBN design incorporates new training stages with approaches to facilitate proper setting of necessary network parameters. The use of training algorithms with minimal heuristics allows the new RBN design to produce results with quality that none of the competing methods have achieved. The new RBN design, called Opt_RBN, is tested on experimental and practical problems, and the results outperform those produced from standard regression and ANN models. In general, the Opt_RBN shows stable and robust performance for a given set of training cases. When the Opt_RBN is applied on the large-scale motion prediction application, the network experiences a CPU memory issue when performing the optimization step in the training process. Therefore, new algorithms are introduced to modify some steps of the new Opt_RBN training process to address the memory issue. The modified steps should only be used for large-scale applications similar to the motion problem. The new RBN design proposes an ANN that is capable of improved learning without needing more training data. Although the new design is driven by its use with motion prediction problems, the consequent ANN design can be used with a broad range of large-scale problems in various engineering and industrial fields that experience delay issues when running computational tools that require a massive number of procedures and a great deal of CPU memory. The results of evaluating the modified Opt_RBN design on two motion problems are promising, with relatively small errors obtained when predicting approximately 500-700 outputs. In addition, new methods for constraint implementation within the new RBN design are introduced. Moreover, the new RBN design and its associated parameters are used as a tool for simulated task analysis. This work initiates the idea that output weights (W) can be used to determine the most critical basis functions that cause the greatest reduction in the network test error. Then, the critical basis functions can specify the most significant training cases that are responsible for the proper performance achieved by the network. The inputs with the most change in value can be extracted from the basis function centers (U) in order to determine the dominant inputs. The outputs with the most change in value and their corresponding key body degrees-of-freedom for a motion task can also be specified using the training cases that are used to create the network's basis functions.

This work is to investigate the performance of two Kalman Filter Algorithms, namely Linear Kalman Filter and Extended Kalman Filter on control-based human motion prediction in a real-time teleoperation. The Kalman Filter Algorithm has been widely used in research areas of motion tracking and GPS-navigation. However, the potential of human motion prediction by utilizing this algorithm is rarely being mentioned. Combine with the known issue - the delay issue in today’s teleoperation services, the author decided to build a prototype of simple teleoperation model based on the Kalman Filter Algorithm with the aim of eliminated the unsynchronization between the user’s inputs and the visual frames, where all the data were transferred over the network. In the first part of the thesis, two types of Kalman Filter Algorithm are applied on the prototype to predict the movement of the robotic arm based on the user’s motion applied on a Haptic Device. The comparisons in performance among the Kalman Filters have also been focused. In the second part, the thesis focuses on optimizing the motion prediction which based on the results of Kalman filtering by using the smoothing algorithm. The last part of the thesis examines the limitation of the prototype, such as how much the delays are accepted and how fast the movement speed of the Phantom Haptic can be, to still be able to obtain reasonable predations with acceptable error rate.   The results show that the Extended Kalman Filter has achieved more advantages in motion prediction than the Linear Kalman Filter during the experiments. The unsynchronization issue has been effectively improved by applying the Kalman Filter Algorithm on both state and measurement models when the latency is set to below 200 milliseconds. The additional smoothing algorithm further increases the accuracy. More important, it also solves shaking issue on the visual frames on robotic arm which is caused by the wavy property of the Kalman Filter Algorithm. Furthermore, the optimization method effectively synchronizes the timing when robotic arm touches the interactable object in the prediction.   The method which is utilized in this research can be a good reference for the future researches in control-based human motion tracking and prediction. / Detta arbete fokuserar på att undersöka prestandan hos två Kalman Filter Algoritmer, nämligen Linear Kalman Filter och Extended Kalman Filter som används i realtids uppskattningar av kontrollbaserad mänsklig rörelse i teleoperationen. Dessa Kalman Filter Algoritmer har används i stor utsträckning forskningsområden i rörelsespårning och GPS-navigering. Emellertid är potentialen i uppskattning av mänsklig rörelse genom att utnyttja denna algoritm sällan nämnas. Genom att kombinera med det kända problemet – fördröjningsproblem i dagens teleoperation tjänster beslutar författaren att bygga en prototyp av en enkel teleoperation modell vilket är baserad på Kalman Filter algoritmen i syftet att eliminera icke-synkronisering mellan användarens inmatningssignaler och visuella information, där alla data överfördes via nätverket. I den första delen av avhandlingen appliceras både Kalman Filter Algoritmer på prototypen för att uppskatta rörelsen av robotarmen baserat på användarens rörelse som anbringas på en haptik enhet. Jämförelserna i prestandan bland de Kalman Filter Algoritmerna har också fokuserats. I den andra delen fokuserar avhandlingen på att optimera uppskattningar av rörelsen som baserat på resultaten av Kalman-filtrering med hjälp av en utjämningsalgoritm. Den sista delen av avhandlingen undersökes begräsning av prototypen, som till exempel hur mycket fördröjningar accepteras och hur snabbt den haptik enheten kan vara, för att kunna erhålla skäliga uppskattningar med acceptabel felfrekvens.   Resultaten visar att den Extended Kalman Filter har bättre prestandan i rörelse uppskattningarna än den Linear Kalman Filter under experimenten. Det icke-synkroniseringsproblemet har förbättrats genom att tillämpa de Kalman Filter Algoritmerna på både statliga och värderingsmodeller när latensen är inställd på under 200 millisekunder. Den extra utjämningsalgoritmen ökar ytterligare noggrannheten. Denna algoritm löser också det skakande problem hos de visuella bilder på robotarmen som orsakas av den vågiga egenskapen hos Kalman Filter Algoritmen. Dessutom effektivt synkroniserar den optimeringsmetoden tidpunkten när robotarmen berör objekten i uppskattningarna.   Den metod som används i denna forskning kan vara en god referens för framtida undersökningar i kontrollbaserad rörelse- spåning och uppskattning.