Nitendra Nath

A Neural Network Controller for Continuum Robots

Continuum or hyper-redundant robot manipulators can exhibit behavior similar to biological trunks, tentacles, or snakes. Unlike traditional rigid-link robot manipulators, continuum robot manipulators do not have rigid joints, hence these manipulators are extremely dexterous, compliant, and are capable of dynamic adaptive manipulation in unstructured environments. However, the development of high-performance control algorithms for these manipulators is quite a challenge, due to their unique design and the high degree of uncertainty in their dynamic models. In this paper, a controller for continuum robots, which utilizes a neural network feedforward component to compensate for dynamic uncertainties is presented. Experimental results using the OCTARM, which is a soft extensible continuum manipulator, are provided to illustrate that the addition of the neural network feedforward component to the controller provides improved performance.

Teleoperation with kinematically redundant robot manipulators with sub-task objectives*

In this paper, control of nonlinear teleoperator systems where both the master and slave systems are kinematically redundant robot manipulators is addressed. The controller is developed under the assumption that the user and environmental input forces are unmeasurable. Lyapunov-based stability analysis is used to prove that the proposed controller yields asymptotic tracking results and ensures the coordination of the master and slave systems while satisfying a sub-task objective. Numerical simulation results are presented to illustrate the effectiveness of the proposed controller.

Neural Network Grasping Controller for Continuum Robots

Continuum or hyper-redundant robots are robots which exhibit behavior similar to biological trunks, tentacles and snakes. Unlike traditional robots, continuum robot manipulators do not have rigid joints, hence the manipulators are compliant, extremely dexterous, and capable of dynamic, adaptive manipulation in unstructured environments; however, the development of high-performance control algorithms for these manipulators is a challenging problem. In this paper, we present an approach to whole arm grasping control for continuum robots. The grasping controller is developed in two stages; high level path planning for the grasping objective, and a low level joint controller using a neural network feedforward component to compensate for dynamic uncertainties. These techniques are used to enable whole arm grasping without using contact force measurements and without using a dynamic model of the continuum robot

Optimizing antiangiogenic therapy for tumor minimization

In this paper, optimization of antiangiogenic therapy for tumor management is considered as a nonlinear control problem. A new technique is developed to optimize antiangiogenic therapy which minimizes the volume of a tumor and prevents it from growing using an optimum drug dose. To this end, an optimum desired trajectory is designed to minimize a performance index. Two controllers are then presented that drive the tumor volume to its optimum value. The first controller is proven to yield exponential results given exact model knowledge. The second controller is developed under the assumption of parameteric uncertainties in the system model. A least-squares estimation strategy based on a prediction error formulation and a Lyapunov-type stability analysis is developed to estimate the unknown parameters of the performance index. An adaptive controller is then designed to track the desired optimum trajectory. The proposed tumor minimization scheme is shown to minimize the tumor volume with an optimum drug dose despite the lack of knowledge of system parameters.

Range identification for nonlinear parameterizable paracatadioptric systems

In this paper, a new range identification technique for a calibrated paracatadioptric system mounted on a moving platform is developed to recover the range information and the three-dimensional (3D) Euclidean coordinates of a static object feature. The position of the moving platform is assumed to be measurable. To identify the unknown range, first a function of the projected pixel coordinates is related to the unknown 3D Euclidean coordinates of an object feature. This function is nonlinearly parameterized (i.e., the unknown parameters appear nonlinearly in the parameterized model). An adaptive estimator based on a min-max algorithm is then designed to estimate the unknown 3D Euclidean coordinates of an object feature relative to a fixed reference frame which facilitates the identification of range. A Lyapunov-type stability analysis is used to show that the developed estimator provides an estimation of the unknown parameters within a desired precision.

Teleoperation with kinematically redundant robot manipulators with sub-task objectives

In this paper, control of nonlinear teleoperator systems where both the master and slave systems are kinematically redundant robot manipulators is addressed. The controller is developed under the assumption that the user and environmental input forces are unmeasurable. Lyapunov-based stability analysis is used to prove that the proposed controller yields asymptotic tracking results and ensures the coordination of the master and slave systems while satisfying a sub-task objective.

Lyapunov-based continuous-stirred tank bioreactor control to maximize biomass production using the haldane and monod specific growth models

A novel robust controller is proposed for a continuous-stirred tank bioreactor that controls the culture dilution rate into the bioreactor in order to maximize a cost function representing the biomass yield. To that end, an optimal desired biomass concentration trajectory is designed based on a numerical extremum-seeking algorithm to maximize the biomass yield. A nonlinear robust controller is designed to ensure the biomass concentration tracks the desired trajectory while providing stable operation. Lyapunov-based stability analyses are used to prove semi-global tracking.

Position based structure from motion using a moving calibrated camera

In this paper, a 3D Euclidean position estimator using a single moving calibrated camera whose position is known is developed that asymptotically recovers the structure of a static object. To estimate the unknown structure an adaptive least squares estimation strategy is employed based on a novel prediction error formulation and a Lyapunov stability analysis. Numerical simulation results are presented to illustrate the effectiveness of the proposed algorithm.

Euclidean Position Estimation of Static Features Using a Moving Uncalibrated Camera

In this paper, a novel Euclidean position estimation technique using a single uncalibrated camera mounted on a moving platform is developed to asymptotically recover the 3-D Euclidean position of static object features. The position of the moving platform is assumed to be measurable, and a second object with known 3-D Euclidean coordinates relative to the world frame is considered to be available a priori. To account for the unknown camera calibration parameters and to estimate the unknown 3-D Euclidean coordinates, an adaptive least squares estimation strategy is employed based on prediction error formulations and a Lyapunov-type stability analysis. The developed estimator is shown to recover the 3-D Euclidean position of the unknown object features despite the lack of knowledge of the camera calibration parameters. Numerical simulation results along with experimental results are presented to illustrate the effectiveness of the proposed algorithm.

RANGE IDENTIFICATION FOR PERSPECTIVE VISION SYSTEMS: A POSITION-BASED APPROACH

In this paper, a new estimator using a single moving calibrated camera is developed to asymptotically recover the range (depth) and the 3D Euclidean position of a static object feature. The position and the orientation of the camera is assumed to be measurable unlike in existing observers where velocity measurements are assumed to be known. To estimate the unknown depth along with the 3D coordinate of a feature an adaptive least squares estimation strategy is employed based on a novel prediction error formulation and a Lyapunov stability analysis. The developed estimator has a simple mathematical structure, can be used to identify range and 3D Euclidean coordinates of multiple features, and is easy to implement. Numerical simulation results along with experimental results are presented to illustrate the effectiveness of the proposed algorithm.