Molecular biological systems can be seen as extremely complex mobile systems. The development of methods for modeling the structure and the motion of such systems is essential to better understand their physiochemical properties and biological functions. In recent years, many computer scientists in robotics and artificial intelligence have made significant contributions to modeling biological systems. Research expertise in planning, search, learning, evolutionary computation, constraint programming, and data mining is being used to make great progress on molecular motion, structure prediction, and design.

Molecular replacement (MR) is a well-established computational method for phasing in macromolecular crystallography. In MR searches, spaces of motions are explored for determining the appropriate placement of rigid single-body (or articulated multi-rigid-body) models of macromolecules. By determining a priori which portions of motion space correspond to non-physical packing arrangements with symmetry mates in collision, it becomes feasible to construct more efficient MR techniques which avoid searching in these non-realizable regions of motion space. This paper investigates which portion of the motion space is physically realizable, given that packing of protein molecules in a crystal are subject to the constraint that they cannot interpenetrate, and gives explicit expressions for the volume of the non-realizable regions for crystals in two-dimensions.

Evidence is emerging that the role of protein structure in disease needs to be rethought. Sequence mutations in proteins are often found to affect the rate at which a protein switches between structures. Modeling structural transitions in wildtype and variant proteins is central to understanding the molecular basis of disease. This paper investigates an efficient algorithmic realization of the stochastic roadmap simulation framework to model structural transitions in wildtype and variants of proteins implicated in human disorders. Our results indicate that the algorithm is able to extract useful information on the impact of mutations on protein structure and function.

We present two methods to find all the possible conformations of short six degree-of-freedom segments of biopolymers which satisfy end constraints in position and orientation. One of our methods is motivated by inverse kinematic solution techniques which have been developed for “general” 6R serial robotic manipulators. However, conventional robot kinematics methods are not directly applicable to the geometry of polymers, which can be treated as a degenerate case where all the “link lengths” are zero. Here, we propose a method which extends the elimination method of Kohli and Osvatic. This method can be applied directly to the geometry of biopolymers. We also propose a heuristic method based on a Lie-group-theoretic description. In this method, we utilize inverse iterations of the Jacobian matrix to obtain all conformations which satisfy end constraints. This can be easily implemented for both the general 6R manipulator and polymers. Although the extended elimination method is computationally faster than the Jacobian method, in cases where some of the joint angles are 180° (i.e., where the elimination method fails), we combine these two methods effectively to obtain the full set of inverse kinematic solutions. We demonstrate our approach with several numerical examples.

It is estimated that allergies afflict up to 40% of the world's population. A primary mediator for allergies is the aggregation of antigens and IgE antibodies bound to cell-surface receptors, FcεRI. Antibody/antigen aggregate formation causes stimulation of mast cells and basophils, initiating cellular degranulation and releasing immune mediators which produce an allergic or anaphylactic response. Understanding the shape and structure of these aggregates can provide critical insights into the allergic response. We have previously developed methods to geometrically model, simulate and analyze antibody aggregation inspired by rigid body robotic motion simulations. Our technique handles the large size and number of molecules involved in aggregation, providing an advantage over traditional simulations such as molecular dynamics (MD) and coarse-grained energetic models. In this paper, we study the impact of model resolution on simulations of geometric structures using both our previously developed Monte Carlo simulation and a novel application of rule-based modeling. These methods complement each other, the former providing explicit geometric detail and the latter providing a generic representation where multiple resolutions can be captured. Our exploration is focused on two antigens, a man-made antigen with three binding sites, DF3, and a common shrimp allergen (antigen), Pen a 1. We find that impact of resolution is minimal for DF3, a small globular antigen, but has a larger impact on Pen a 1, a rod-shaped molecule. The volume reduction caused by the loss in resolution allows more binding site accessibility, which can be quantified using a rule-based model with implicit geometric input. Clustering analysis of our simulation shows good correlation when compared with available experimental results. Moreover, collisions in all-atom reconstructions are negligible, at around 0.2% at 90% reduction.

The cyclic coordinate descent (CCD) method is a popular loop closure method in protein structure modeling. It is a robotics algorithm originally developed for inverse kinematic applications. We demonstrate an effective method of building the backbone of protein structure models using the principle of CCD and a guiding trace. For medium-resolution 3-dimensional (3D) images derived using cryo-electron microscopy (cryo-EM), it is possible to obtain guiding traces of secondary structures and their skeleton connections. Our new method, constrained cyclic coordinate descent (CCCD), builds α-helices, β-strands, and loops quickly and fairly accurately along predefined traces. We show that it is possible to build the entire backbone of a protein fairly accurately when the guiding traces are accurate. In a test of 10 proteins, the models constructed using CCCD show an average of 3.91 Å of backbone root mean square deviation (RMSD). When the CCCD method is incorporated in a simulated annealing framework to sample possible shift, translation, and rotation freedom, the models built with the true topology were ranked high on the list, with an average backbone RMSD100 of 3.76 Å. CCCD is an effective method for modeling atomic structures after secondary structure traces and skeletons are extracted from 3D cryo-EM images.

Detection of targets distributed randomly in space is a task common to both robotic and biological systems. Lévy search has previously been used to characterize T cell search in the immune system. We use a robot swarm to evaluate the effectiveness of a Lévy search strategy and map the relationship between search parameters and target configurations. We show that the fractal dimension of the Lévy search which optimizes search efficiency depends strongly on the distribution of targets but only weakly on the number of agents involved in search. Lévy search can therefore be tuned to the target configuration while also being scalable. Implementing search behaviors observed in T cells in a robot swarm provides an effective, adaptable, and scalable swarm robotic search strategy. Additionally, the adaptability and scalability of Lévy search may explain why Lévy-like movement has been observed in T cells in multiple immunological contexts.

We present an optimal gain scheduling control design for bipedal walking with minimum tracking error. We obtained a linear approximation by linearizing the nonlinear hybrid dynamic model about a nominal periodic trajectory. This linearization allows us to identify the linear model as a linear periodic system. An optimal feedback was designed using Bellman's dynamic programming. The linear periodic system allows us to determine a linear quadratic regulator (LQR) for a single period and to set the Hamilton-Jacobi-Bellman (HJB) function in a linear quadratic form. In this way, the dynamic programming yielded an admissible continuous gain scheduling that was designed with regard to the hybrid dynamics of the system. We tuned the optimization parameters such that the tracking error and the average energy consumption are minimized. Due to linearization, we were able to examine the stability of the approximated periodic system achieved by the periodic gain according to Floquet's theory, by calculating the monodromy matrix of the closed-loop hybrid system. In addition to determining stability, the eigenvalues of this approximated monodromy matrix allowed us to evaluate the settling time of the system. This approach presents a direct method for optimal solution of locomotion control according to a given reference trajectory.

An optimization model is developed in this paper for the joint trajectories of a hopping robot with a four-bar linkage leg. The dynamic behaviour of the one-legged robot is investigated during the stance and swing phases, and their impacts on gait planning are analysed. Certain constraints characterizing the continuous and cyclic motion of the robot are obtained. The optimization model is solved for the minimum torque and maximum velocity objective functions separately, and the results are compared with those in nature.

A centipede-like robot, which has high degrees of freedom (DOFs) and similar body segments, requires new real-time control to achieve diverse gaits. Therefore, we have studied the movement characteristics of multi-legged creatures and determined the features for a gait periodic relay. An instruction-relay control scheme, the three-bus control system and the required software were then designed. In our experiments with the designed control system, different gaits for the robot could be achieved, and the phase difference between body segments could be changed by altering the delay time. As a whole, this control system could accomplish the required control task and reasonably simplify the gait control algorithms and procedures.

This paper presents a new approach to accurately track a moving vehicle with a multiview setup of red–green–blue depth (RGBD) cameras. We first propose a correction method to eliminate a shift, which occurs in depth sensors when they become worn. This issue could not be otherwise corrected with the ordinary calibration procedure. Next, we present a sensor-wise filtering system to correct for an unknown vehicle motion. A data fusion algorithm is then used to optimally merge the sensor-wise estimated trajectories. We implement most parts of our solution in the graphic processor. Hence, the whole system is able to operate at up to 25 frames per second with a configuration of five cameras. Test results show the accuracy we achieved and the robustness of our solution to overcome uncertainties in the measurements and the modelling.

We present a novel motorized semi-autonomous mobile hospital bed guided by a human operator and a reactive navigation algorithm. The proposed reactive navigation algorithm is launched when the sensory device detects that the hospital bed is in the potential danger of collision. The semi-autonomous hospital bed is able to safely and quickly deliver critical neurosurgery (head trauma) patients to target locations in dynamic uncertain hospital environments such as crowded hospital corridors while avoiding en-route steady and moving obstacles. We do not restrict the nature or the motion of the obstacles, meaning that the shapes of the obstacles may be time-varying or deforming and they may undergo arbitrary motions. The only information available to the navigation system is the current distance to the nearest obstacle. Performance of the proposed navigation algorithm is verified via theoretical studies. Simulation and experimental results also confirm the performance of the reactive navigation algorithm in real world scenarios.

This paper deals with the analytical modeling and the analysis of the orientation error of the RAF translator due to the clearances in the joints. This model presents the orientation error as a function of the nominal pose, the external load applied to the platform, the manipulator structural parameters, and the joints clearances. Based on this model, an algorithm is developed in order to map the pose error within a desired workspace of the manipulator. It is shown that the orientation error variation depends essentially on the parallelogram configuration of the passive legs out of its plane. The orientation error magnitude is mainly caused by the parallelogram revolute joints radial clearances. Moreover, the orientation error around the z-axis presents some discontinuities due to the contact mode change of the parallelogram revolute joints.