Shivashankar B. Nair

Bio-inspired artificial intelligence

The lecture provides a fast forward on some of the major fields of bio-inspired AI and concentrate on some recent work carried out on the realization of an Artificial Being at the Robotics Lab. of the Department of Computer Science & Engineering, Indian Institute of Technology Guwahati, India. Mechanisms for patrolling and servicing by mobile agents within a network using bio-inspired techniques and attempts to model emotional robots would also be dealt with. A mobile agent framework nicknamed Typhon to emulate such bio-inspired AI techniques will also be introduced followed by some live and maybe hands-on demos.

Rolling element bearing design through genetic algorithms

The design of rolling element bearings has been a challenging task in the field of mechanical engineering. While most of the real aspects of the design are never disclosed by bearing manufacturers, the common engineer is left with no other alternative than to refer to standard tables and charts containing the bearing performance characteristics. This paper presents a more viable method to solve this problem using genetic algorithms (GAs). Since the algorithm is basically a guided random search, it weakens the chances of getting trapped in local maxima or minima. The method used has yielded improved performance parameters than those catalogued in standard tables. *Indraneel Chakraborty is currently with Massachusetts Institute of Technology and can be reached at indranil@mit.edu †Vinay Kumar is currently with Mindtree Consulting, India and can be reached at vinayk@mindtree.com

An Immune System Based Multi-robot Mobile Agent Network

A biological immune system comprises components that circulate within the body and defend it from attacks by pathogens. Its artificial counterpart has found applications in a wide spectrum of domains including Robotics. The cells in an immune system can be viewed to be analogous to mobile agents that migrate from one node (organ) to another. Mobility of such agents provides the framework for movement of components within a system. In this paper we describe the modeling of a mobile agent based robotic system that works on the principles of an artificial immune system. The mobile agents form the B-cell equivalents, generating rule-sets (antibodies) that help the robot to overcome a problem (antigen).

Typhon : a mobile agents framework for real world emulation in prolog

In this paper, we present a mobile agent framework nicknamed Typhon , based on LPA Prologs Chimera agent system, which allows users to go in for rapid emulation of algorithms and test beds rather than their mere simulation. This framework provides for agent migration, cloning, payload carrying abilities, state saving and security and also facilitates a reduced deployment time. The framework allows users to exploit the intelligence programming abilities of Prolog and integrate them with the inherent parallelism exhibited by mobile agents. To portray the versatility of the system we describe implementations of a typical resource discovery application for a robot tethered to a node in a network.

On Realizing a Multi-Agent Emotion Engine

Emotions have always been a complex phenomenon and research on their causes and effects have been fraught with debates. Though a reasonable and unified theory seems lacking, there have been many attempts at building models that emote. This paper describes a multi-agent approach that aids robot emotion. Emotions are grounded on percepts from sensors and generated by dedicated emotion agents that work concurrently with others – the positive suppressing the negative and vice versa while stimulating their own kinds. Each agent forms a metaphor of an emotion-generating entity that has a replenishing capability. Both the replenishing of an emotion resource and the sampling of the environment are based on fuzzy logic. Sampling of the percepts from the sensors is based on an adrenaline-like effect. Stimulations, suppressions, emotion resource, and a look-back before decay feature embed a deep and dynamic emotional milieu into a machine. The paper presents and discusses how three emotions churned from percepts gathered by a robot act as an emotional control juice capable of governing the manner of its motion along a path.

On a Mobile Agent Framework for an Internet of Things

With a large number of heterogeneous devices and gadgets ready to populate the Internet, finding an effective mechanism for communication amongst them has become imperative. Communication between networked heterogeneous devices can allow them to utilize each others individual capabilities to perform a task co-operatively. We propose the use of mobile agents to not only achieve such communication but also to pro-actively search for resources and provide services to the various devices populating such a network. In this paper we describe a mobile agent framework for an Internet of Things and discuss its working through a real implementation.

Towards a dynamic emotional model

Emotions in general are a complex phenomenon to comprehend. They have been known to influence decisions as also behavior. Several computational models for emotions have been attempted. In this paper we present the dynamics and working of a multi-agent emotion generating system. Three basic emotions viz. happiness, fear and anger have been used to generate a resultant internal mood for a machine or a robot whose state depends on the immediate past. We compare the results obtained with those generated by a conventional fuzzy model to highlight their significance.

Mobile agent cloning for servicing networked robots

In this work we present how the concept of cloning of mobile agents can aid in enhancing the performance of a multi robot system by providing faster services to the robots. The mobile agents carry solutions to problems faced by robots, as their payload. These agents move conscientiously in the network till they sense a pheromone gradient formed by a Robot Requesting Service (RRS) and its neighbours. Mobile agents tend to clone when they discover pheromones diffused by different RRSs thereby resulting in a faster and concurrent service. Simulation results have shown that cloning agents in high density RRS areas greatly decreases the waiting times for a service.

An artificial immune system based approach for English grammar checking

Grammar checking and correction comprise of the primary problems in the area of Natural Language Processing (NLP). Traditional approaches fall into two major categories: Rule based and Corpus based. While the former relies heavily on grammar rules the latter approach is statistical in nature. We provide a novel corpus based approach for grammar checking that uses the principles of an Artificial Immune System (AIS).We treat grammatical error as pathogens (in immunological terms) and build antibody detectors capable of detecting grammatical errors while allowing correct constructs to filter through. Our results show that it is possible to detect a range of grammatical errors. This method can prove extremely useful in applications like Intelligent Tutoring Systems (ITS) and general purpose grammar checkers.

Design and implementation of a spell checker for Assamese

Spell checkers form a vital ingredient of text processors, character recognition systems, dictionary search engines, language processing software and similar tools. Although considerable work has been done in the area for English and related languages, the Indian language scenario presents a relatively more complex and uphill task. This paper describes strategies involved in the implementation of a spell checker for Assamese, the official language or the North Eastern Indian State of Assam.