Xianlian Zhou

Towards a new generation of virtual humans

This paper presents work from an ongoing project towards developing a new generation of virtual human models that are highly realistic in appearance, movement, and feedback. Santos™, an anatomically correct human model with more than 100 degrees of freedom, is an avatar that exhibits extensive modelling and simulation capabilities, resides in a virtual environment, and conducts human-factors analysis. The paper presents an optimisation-based approach to posture and motion prediction that allows the avatar to operate with autonomy rather than depending on stored animations and data or being restricted by inverse kinematics. It also presents approaches to determining reach envelopes and workspace zone differentiation, and discusses methods for evaluating the physiological status of the virtual human as it completes tasks. Muscle modelling including muscle wrapping, muscle force and stress determination is also discussed. Finally, the process of building a 25-DOF hand model is described. The result is an exciting step towards a virtual human that is more extensive and complete than any other.

NURBS-based Galerkin method and application to skeletal muscle modeling

Non-Uniform Rational B-spline (NURBS) is often used to construct the free-form boundary representation of three-dimensional objects. In this paper, we propose a method for mechanical analysis for deformable bodies by combining NURBS geometric representation and the Galerkin method. The NURBS surface bounding a 3D body is extended to a trivariate NURBS solid by adding another parametric domain represented by additional control points. The displacement field of the body is constructed using the NURBS shape representation with the control point being the generalized coordinates. The interpolated displacement field is directly used to facilitate finite element formulation. In this manner, traditional FEM meshing is not required. In this work, the NURBS-FEM is applied to skeletal muscle modeling. Muscle is modeled as anisotropic, active hyperelastic solids. The directions of the contractile fibers can be uniform or along the tangent direction of NURBS curves. Typical contractive motions of isolated muscle are simulated.

Patient-Specific Wall Stress Analysis in Cerebral Aneurysms Using Inverse Shell Model

Stress analyses of patient-specific vascular structures commonly assume that the reconstructed in vivo configuration is stress free although it is in a pre-deformed state. We submit that this assumption can be obviated using an inverse approach, thus increasing accuracy of stress estimates. In this paper, we introduce an inverse approach of stress analysis for cerebral aneurysms modeled as nonlinear thin shell structures, and demonstrate the method using a patient-specific aneurysm. A lesion surface derived from medical images, which corresponds to the deformed configuration under the arterial pressure, is taken as the input. The wall stress in the given deformed configuration, together with the unstressed initial configuration, are predicted by solving the equilibrium equations as opposed to traditional approach where the deformed geometry is assumed stress free. This inverse approach also possesses a unique advantage, that is, for some lesions it enables us to predict the wall stress without accurate knowledge of the wall elastic property. In this study, we also investigate the sensitivity of the wall stress to material parameters. It is found that the in-plane component of the wall stress is indeed insensitive to the material model.

Estimation of vascular open configuration using finite element inverse elastostatic method

This paper presents a new method for predicting the open configuration of vascular organs. The method utilizes finite element inverse elastostatic formulations. The equilibrium boundary value problem is formulated on the homeostatic configuration, and is solved inversely to find the open, stress-free configuration. The method is non-invasive, and enables us to estimate the open configuration based on information that is readily available form in vivo measurements. Examples involving both axisymmetric and asymmetric geometries are presented to demonstrate the utility of the method.

Cortical Spiking Network Interfaced with Virtual Musculoskeletal Arm and Robotic Arm.

Embedding computational models in the physical world is a critical step towards constraining their behavior and building practical applications. Here we aim to drive a realistic musculoskeletal arm model using a biomimetic cortical spiking model, and make a robot arm reproduce the same trajectories in real time. Our cortical model consisted of a 3-layered cortex, composed of several hundred spiking model-neurons, which display physiologically realistic dynamics. We interconnected the cortical model to a two-joint musculoskeletal model of a human arm, with realistic anatomical and biomechanical properties. The virtual arm received muscle excitations from the neuronal model, and fed back proprioceptive information, forming a closed-loop system. The cortical model was trained using spike timing-dependent reinforcement learning to drive the virtual arm in a 2D reaching task. Limb position was used to simultaneously control a robot arm using an improved network interface. Virtual arm muscle activations responded to motoneuron firing rates, with virtual arm muscles lengths encoded via population coding in the proprioceptive population. After training, the virtual arm performed reaching movements which were smoother and more realistic than those obtained using a simplistic arm model. This system provided access to both spiking network properties and to arm biophysical properties, including muscle forces. The use of a musculoskeletal virtual arm and the improved control system allowed the robot arm to perform movements which were smoother than those reported in our previous paper using a simplistic arm. This work provides a novel approach consisting of bidirectionally connecting a cortical model to a realistic virtual arm, and using the system output to drive a robotic arm in real time. Our techniques are applicable to the future development of brain neuroprosthetic control systems, and may enable enhanced brain-machine interfaces with the possibility for finer control of limb prosthetics.

Virtual musculoskeletal arm and robotic arm driven by a biomimetic model of sensorimotor cortex with reinforcement learning

Neocortical mechanisms of learning sensorimotor control involve a complex series of interactions at multiple levels, from synaptic mechanisms to network connectomics. We developed a model of sensory and motor cortex consisting of several hundred spiking model-neurons. A biomimetic model (BMM) was trained using spike-timing dependent reinforcement learning to drive a simple kinematic two-joint virtual arm in a motor task requiring convergence on a single target. After learning, networks demonstrated retention of behaviorally-relevant memories by utilizing proprioceptive information to perform reach-to-target from multiple starting positions. We utilized the output of this model to drive mirroring motion of a robotic arm. In order to improve the biological realism of the motor control system, we replaced the simple virtual arm model with a realistic virtual musculoskeletal arm which was interposed between the BMM and the robot arm. The virtual musculoskeletal arm received input from the BMM signaling neural excitation for each muscle. It then fed back realistic proprioceptive information, including muscle fiber length and joint angles, which were employed in the reinforcement learning process. The limb position information was also used to control the robotic arm, leading to more realistic movements. This work explores the use of reinforcement learning in a spiking model of sensorimotor cortex and how this is affected by the bidirectional interaction with the kinematics and dynamic constraints of a realistic musculoskeletal arm model. It also paves the way towards a full closed-loop biomimetic brain-effector system that can be incorporated in a neural decoder for prosthetic control, and used for developing biomimetic learning algorithms for controlling real-time devices. Additionally, utilizing biomimetic neuronal modeling in brain-machine interfaces offers the possibility for finer control of prosthetics, and the ability to better understand the brain.

A fast and robust whole-body control algorithm for running

Dynamic simulation of human motion in real-time is a challenging problem due to the high mobility of human body and the redundancy of body control. This paper presents a robust whole-body control algorithm for simulating human running in real-time. The control algorithm tracks a reference motion through a proportional derivative (PD) feedback control rule for computing desired joint accelerations and an efficient and novel procedure for predicting ground reaction force (GRF) and joint actuation torques, which are subsequently applied to a forward dynamics simulation. The proposed control algorithm is firstly demonstrated to track a measured running motion and the predicted GRF is shown in good agreement with the experimental data. Further, the capability to handle extra loads is demonstrated with load carriage studies. In summary, the present approach offers a fast and robust way to synthesise physically valid dynamic motions from captured motion samples without the need of measured GRF.

Anthropometry model generation based on ANSUR II database

This paper presents a three-dimensional (3D) anthropometry model generation (AMG) software framework based on the latest US Army Anthropometric Survey (ANSUR II) database. The software utilises principal component shapes derived from 3D body scans in the database and employs two complementary methods (feature analysis method and database method) to generate 3D surface models directly from anthropometric feature specifications. Virtual landmarks and 3D features identified on the surface models are the keys for success of the present approaches. It is demonstrated that the features calculated from these models are in good agreement with traditional tape or caliper measurements with a few exceptions due to measurement methodology or posture. Besides surface anthropometry, the software is able to construct a dynamic joint segment framework from any of the models and morph associated bones, muscles, interior organs, and clothing accordingly. This AMG software is a valuable tool for AMG in digital human/warfighter/patient modelling for various applications in biomechanics, human factors or ergonomics, physiology, and medicine.

A musculoskeletal fatigue model for prediction of aviator neck manoeuvring loadings

A new musculoskeletal fatigue model was developed and demonstrated on the prediction of neck musculoskeletal loading and fatigue of fighter pilots during high-G aerial combat manoeuvring (ACM). A whole body articulated multi-body model with detailed neck musculature, representing a 50th percentile male, was utilised in this study. To account for the decrease of muscle force generation capacity during intensive muscle contraction, a new dynamic muscle fatigue model based on a fatigue-rest-recovery mechanism was incorporated to predict muscle fatigue responding to arbitrary neural excitations. Two flight postures, look-ahead and check-6, were investigated on their effects on neck loading during a 6-minute idealised ACM profile. The joint dynamics, muscle forces and fatigue levels, representing the neck biodynamic responses to the applied aircraft acceleration, were obtained and compared for these two postures. Significant differences are observed which show the check-6 posture poses much greater challenge to muscle force generation and induces quicker and severer muscle fatigue.

Biomechanical analysis of skeletal muscle in an interactive digital human system

Biomechanical analysis of skeletal muscles is an important task in digital human systems. The standard finite element method (FEM) can be used for muscle analysis; however, a full-scale FEM model can be overly complicated in a digital human system. In this work, we describe an efficient method of muscle analysis. The method is a combination of the Non-Uniform Rational B-spline (NURBS) geometric representation and the Galerkin methods. The basic idea is to establish the discrete equations of motion on the basis of NURBS geometry directly, without resorting to additional meshing. The method can adequately model muscle motion and stress while keeping the model size and complexity at a tractable level. As the first step towards interactive stress analysis in a digital human, we have developed NURBS FEM model for isolated muscles in human upper limb. The geometries of the muscles are extracted from the Visible Human Data Set [15], and the mechanical behavior is characterized by an active, anisotropic hyperelastic model. We also discuss how the muscle model will be implemented in and interacts with the virtual human (Santos) developed at the University of Iowa.