Swagat Kumar

On Adaptive Learning Rate That Guarantees Convergence in Feedforward Networks

This paper investigates new learning algorithms (LF I and LF II) based on Lyapunov function for the training of feedforward neural networks. It is observed that such algorithms have interesting parallel with the popular backpropagation (BP) algorithm where the fixed learning rate is replaced by an adaptive learning rate computed using convergence theorem based on Lyapunov stability theory. LF II, a modified version of LF I, has been introduced with an aim to avoid local minima. This modification also helps in improving the convergence speed in some cases. Conditions for achieving global minimum for these kind of algorithms have been studied in detail. The performances of the proposed algorithms are compared with BP algorithm and extended Kalman filtering (EKF) on three bench-mark function approximation problems: XOR, 3-bit parity, and 8-3 encoder. The comparisons are made in terms of number of learning iterations and computational time required for convergence. It is found that the proposed algorithms (LF I and II) are much faster in convergence than other two algorithms to attain same accuracy. Finally, the comparison is made on a complex two-dimensional (2-D) Gabor function and effect of adaptive learning rate for faster convergence is verified. In a nutshell, the investigations made in this paper help us better understand the learning procedure of feedforward neural networks in terms of adaptive learning rate, convergence speed, and local minima

Malabsorption of Rifampin and Isoniazid in HIV-Infected Patients With and Without Tuberculosis

The absorption of rifampin, isoniazid, and D-xylose in patients with human immunodeficiency virus (HIV) infection and diarrhea, in patients with HIV infection and tuberculosis (TB), in patients with pulmonary TB alone, and in healthy subjects was studied. Percentage of dose of the drugs, their metabolites, and D-xylose excreted in urine were calculated. A significant reduction in the absorption of drugs and D-xylose in both the HIV infection/diarrhea and HIV infection/TB groups was observed (P<.05), and the correlation between them was significant. Our results indicate that patients with HIV infection and diarrhea and those with HIV infection and TB have malabsorption of rifampin and isoniazid.

Kinematic control of a redundant manipulator using an inverse-forward adaptive scheme with a KSOM based hint generator

This paper proposes an online inverse-forward adaptive scheme with a KSOM based hint generator for solving the inverse kinematic problem of a redundant manipulator. In this approach, a feed-forward network such as a radial basis function (RBF) network is used to learn the forward kinematic map of the redundant manipulator. This network is inverted using an inverse-forward adaptive scheme until the network inversion solution guides the manipulator end-effector to reach a given target position with a specified accuracy. The positioning accuracy, attainable by a conventional network inversion scheme, depends on the approximation error present in the forward model. But, an accurate forward map would require a very large size of training data as well as network architecture. The proposed inverse-forward adaptive scheme effectively approximates the forward map around the joint angle vector provided by a hint generator. Thus the inverse kinematic solution obtained using the network inversion approach can take the end-effector to the target position within any arbitrary accuracy. In order to satisfy the joint angle constraints, it is necessary to provide the network inversion algorithm with an initial hint for the joint angle vector. Since a redundant manipulator can reach a given target end-effector position through several joint angle vectors, it is desirable that the hint generator is capable of providing multiple hints. This problem has been addressed by using a Kohonen self organizing map based sub-clustering (KSOM-SC) network architecture. The redundancy resolution process involves selecting a suitable joint angle configuration based on different task related criteria. The simulations and experiments are carried out on a 7 DOF PowerCube(TM) manipulator. It is shown that one can obtain a positioning accuracy of 1 mm without violating joint angle constraints even when the forward approximation error is as large as 4 cm. An obstacle avoidance problem has also been solved to demonstrate the redundancy resolution process with the proposed scheme.

Visual motor control of a 7dof redundant manipulator using redundancy preserving learning network

This paper deals with the design and implementation of a visual kinematic control scheme for a redundant manipulator. The inverse kinematic map for a redundant manipulator is a one-to-many relation problem; i.e. for each Cartesian position, multiple joint angle vectors are associated. When this inverse kinematic relation is learnt using existing learning schemes, a single inverse kinematic solution is achieved, although the manipulator is redundant. Thus a new redundancy preserving network based on the self-organizing map (SOM) has been proposed to learn the one-to-many relation using sub-clustering in joint angle space. The SOM network resolves redundancy using three criteria, namely lazy arm movement, minimum angle norm and minimum condition number of image Jacobian matrix. The proposed scheme is able to guide the manipulator end-effector towards the desired target within 1-mm positioning accuracy without exceeding physical joint angle limits. A new concept of neighbourhood has been introduced to enable the manipulator to follow any continuous trajectory. The proposed scheme has been implemented on a seven-degree-of-freedom (7DOF) PowerCube robot manipulator successfully with visual position feedback only. The positioning accuracy of the redundant manipulator using the proposed scheme outperforms existing SOM-based algorithms.

Visual Motor Control of a 7 DOF Robot Manipulator Using Function Decomposition and Sub-Clustering in Configuration Space

This paper deals with real-time implementation of visual-motor control of a 7 degree of freedom (DOF) robot manipulator using self-organized map (SOM) based learning approach. The robot manipulator considered here is a 7 DOF PowerCube manipulator from Amtec Robotics. The primary objective is to reach a target point in the task space using only a single step movement from any arbitrary initial configuration of the robot manipulator. A new clustering algorithm using Kohonen SOM lattice has been proposed that maintains the fidelity of training data. Two different approaches have been proposed to find an inverse kinematic solution without using any orientation feedback. In the first approach, the inverse Jacobian matrices are learnt from the training data using function decomposition. It is shown that function decomposition leads to significant improvement in accuracy of inverse kinematic solution. In the second approach, a concept called sub-clustering in configuration space is suggested to provide multiple solutions for the inverse kinematic problem. Redundancy is resolved at position level using several criteria. A redundant manipulator is dexterous owing to the availability of multiple configurations for a given end-effector position. However, existing visual motor coordination schemes provide only one inverse kinematic solution for every target position even when the manipulator is kinematically redundant. Thus, the second approach provides a learning architecture that can capture redundancy from the training data. The training data are generated using explicit kinematic model of the combined robot manipulator and camera configuration. The training is carried out off-line and the trained network is used on-line to compute the joint angle vector to reach a target position in a single step only. The accuracy attained is better than the current state of art.

Direct Adaptive Control using Single Network Adaptive Critic

An optimal control law for a general nonlinear system can be obtained by solving Hamilton-Jacobi-Bellman equation. However, it is difficult to obtain an analytical solution of this equation. In this paper, we propose a direct adaptive control method for affine systems where the optimal value function is approximated using a critic network and the weight update law is derived so as to satisfy the HJB equation. This provides an alternative method to solve the HJB equation in real-time and thus paves the way for introducing some notion of optimality to direct adaptive control paradigm. The efficacy of the scheme is demonstrated for a linearized mass-spring-damper system and its performance is compared with that of LQR.

High performance loop closure detection using bag of word pairs

We propose a new method for loop closure detection for topological mapping.It uses relative spatial co-occurrence information to improve the performance.We augment BoW method with a dictionary of spatially co-occurring word pairs.A memory map data structure is used for storing and indexing word pairs.We incorporate best of the existing methods to provide state-of-the-art performance. In this paper, we look into the problem of loop closure detection in topological mapping. The bag of words (BoW) is a popular approach which is fast and easy to implement, but suffers from perceptual aliasing, primarily due to vector quantization. We propose to overcome this limitation by incorporating the spatial co-occurrence information directly into the dictionary itself. This is done by creating an additional dictionary comprising of word pairs, which are formed by using a spatial neighborhood defined based on the scale size of each point feature. Since the word pairs are defined relative to the spatial location of each point feature, they exhibit a directional attribute which is a new finding made in this paper. The proposed approach, called bag of word pairs (BoWP), uses relative spatial co-occurrence of words to overcome the limitations of the conventional BoW methods. Unlike previous methods that use spatial arrangement only as a verification step, the proposed method incorporates spatial information directly into the detection level and thus, influences all stages of decision making. The proposed BoWP method is implemented in an on-line fashion by incorporating some of the popular concepts such as, K-D tree for storing and searching features, Bayesian probabilistic framework for making decisions on loop closures, incremental creation of dictionary and using RANSAC for confirming loop closure for the top candidate. Unlike previous methods, an incremental version of K-D tree implementation is used which prevents rebuilding of tree for every incoming image, thereby reducing the per image computation time considerably. Through experiments on standard datasets it is shown that the proposed methods provide better recall performance than most of the existing methods. This improvement is achieved without making use any geometric information obtained from range sensors or robot odometry. The computational requirements for the algorithm is comparable to that of BoW methods and is shown to be less than the latest state-of-the-art method in this category.

Remote retail monitoring and stock assessment using mobile robots

This paper describes a Virtual Reality (VR) based system for automating data collection and surveying in a retail store using mobile robots. The manpower cost for surveying and monitoring the shelves in retail stores are high, because of which these activities are not repeated frequently causing reduced customer satisfaction and loss of revenue. Further, the accuracy of data collected may be improved by avoiding human related factors. We use a mobile robot platform with on-board cameras to monitor the shelves either autonomously or through tele-operation. A remote operator can control the robot from a console which shows a 3D of view of the store as well as, capture real images and videos of the store. The robot is designed to facilitate automatic detection of Out-of-Stock (OOS) situations. It would be possible for a single operator to control multiple robots placed at different stores thus optimizing the available resources. As the deployment of the proposed system does not require modifying existing infrastructure of the store, the cost of the entire solution is cheaper with shorter return-on-investment (ROI) period.

Continuous-time Single Network Adaptive Critic for Regulator Design of Nonlinear Control Affine Systems

An optimal control law for a general nonlinear system can be obtained by solving Hamilton-Jacobi-Bellman equation. However, it is difficult to obtain an analytical solution of this equation even for a moderately complex system. In this paper, we propose a continuoustime single network adaptive critic scheme for nonlinear control affine systems where the optimal cost-to-go function is approximated using a parametric positive semi-definite function. Unlike earlier approaches, a continuous-time weight update law is derived from the HJB equation. The stability of the system is analysed during the evolution of weights using Lyapunov theory. The effectiveness of the scheme is demonstrated through simulation examples.

A Model-free Redundancy Resolution Technique for Visual Motor Coordination of a 6 DOF robot manipulator

In this paper, visual motor coordination of a 6 DOF robot manipulator is considered. It is difficult to analytically derive inverse kinematic relationships for such manipulators. The problem becomes more challenging owing to the presence of multiple solutions for the inverse-kinematic relationship between robot end-effector position and joint angle vector. Many of the current redundancy resolution techniques necessitate explicit orientation information which cannot be obtained from visual feedback. Hence such techniques cannot be used for visual motor coordination of redundant manipulators. In this paper, it is demonstrated that a feasible inverse kinematic solution may be obtained by using input-output space clustering along with the KSOM algorithm. The method is innovative in the sense that it does not require any orientation information for resolving redundancy and it is model-independent. The efficacy of the proposed method is illustrated through simulations on a 6 DOF PUMA 560 manipulator model.