Goutham Mallapragada

Symbolic Dynamic Filtering and Language Measure for Behavior Identification of Mobile Robots

This paper presents a procedure for behavior identification of mobile robots, which requires limited or no domain knowledge of the underlying process. While the features of robot behavior are extracted by symbolic dynamic filtering of the observed time series, the behavior patterns are classified based on language measure theory. The behavior identification procedure has been experimentally validated on a networked robotic test bed by comparison with commonly used tools, namely, principal component analysis for feature extraction and Bayesian risk analysis for pattern classification.

ν☆: a robot path planning algorithm based on renormalised measure of probabilistic regular languages

This article introduces a novel path planning algorithm, called ν ☆, that reduces the problem of robot path planning to optimisation of a probabilistic finite state automaton. The ν ☆-algorithm makes use of renormalised measure ν of regular languages to plan the optimal path for a specified goal. Although the underlying navigation model is probabilistic, the ν ☆-algorithm yields path plans that can be executed in a deterministic setting with automated optimal trade-off between path length and robustness under dynamic uncertainties. The ν ☆-algorithm has been experimentally validated on Segway Robotic Mobility Platforms in a laboratory environment.

Automated behaviour recognition in mobile robots using symbolic dynamic filtering

This paper introduces a dynamic data-driven method for signature detection in mobile robots. The core concept of the paper is built upon the principles of symbolic dynamic filtering (SDF) that can be used to extract relevant information in complex dynamical systems. The objective here is to identify the robot behaviour in real time as accurately as possible. The proposed method is validated by experimentation on a networked robotics test bed to detect and identify the type and motion profile of the robots under consideration.

Autonomous robot navigation using optimal control of probabilistic regular languages

This paper addresses autonomous intelligent navigation of mobile robotic platforms based on the recently reported algorithms of language-measure-theoretic optimal control. Real-time sensor data and model-based information on the robots motion dynamics are fused to construct a probabilistic finite state automaton model that dynamically computes a time-dependent discrete-event supervisory control policy. The paper also addresses detection and avoidance of livelocks that might occur during execution of the robot navigation algorithm. Performance and robustness of autonomous intelligent navigation under the proposed algorithm have been experimentally validated on Segway RMP robotic platforms in a laboratory environment.

Autonomous Navigation of Mobile Robots Using Optimal Control of Finite State Automata

This paper presents a novel approach to autonomous intelligent navigation of mobile robotic platforms, which is based on the concept of language-theoretic discrete-event supervisory control. The proposed algorithm combines real-time sensor data and model-based information on motion dynamics into a probabilistic finite state automaton model to dynamically compute a time-varying supervisory algorithm. The performance and robustness of the autonomous intelligent navigation algorithm have been experimentally validated on Segway RMP robotic platforms

Tracking Mobile Targets Using Wireless Sensor Networks

Mobile target tracking needs a sensor network able to autonomously adapt to the requirements of the data fusion algorithms. We propose a design architecture for a tracking algorithm in which the sensed data is processed in an abstract space called Information Space and the communication between nodes is modeled as an abstract space called Network Design Space. The two abstract spaces are connected through an interaction interface called InfoNet, that seamlessly translates the messages between the two. The proposed architecture is evaluated using simulations in NS2. The sensory data is collected from a laboratory testbed where a mobile robot (a Segway) moves along various types of trajectories over a pressure sensor network.

Language-measure-based supervisory control of a mobile robot

This paper presents the design, modelling, and supervisory control of a mobile robot based on a signed real measure of its automaton (i.e., discrete-event behavior) language. While the robots dynamic behavior is manipulated in the continuous-time domain via motion control and visual servoing, the mission planning is performed in the discrete-event supervisory control setting. However, unlike the conventional qualitative framework of supervisory control following the Ramadge-Wonham approach that is based on a set of specified constraints, a quantitative approach has been adopted for synthesis of optimal supervisory controllers in robotic scenarios with a language measure being the performance index. The parameters of the language measure are identified via both experimental observations and simulation runs; the results are consistent with each other as well as with other measures. This approach complements the Q-learning method that has been widely used in robotics research to learn primitive behaviors.

DESIGNING A FUSION-DRIVEN SENSOR NETWORK TO SELECTIVELY TRACK MOBILE TARGETS

Sensor networks that can support time-critical operations pose challenging problems for tracking events of interest. We propose an architecture for a sensor network that autonomously adapts in real-time to data fusion requirements so as not to miss events of interest and provides accurate real-time mobile target tracking. In the proposed architecture, the sensed data is processed in an abstract space called Information Space and the communication between nodes is modeled as an abstract space called Network Design Space. The two abstract spaces are connected through an interaction interface called InfoNet, that seamlessly translates the messages between the two. The proposed architecture is validated experimentally on a laboratory testbed for multiple scenarios.

Behavior recognition in mobile robots using Symbolic Dynamic Filtering and language measure

This paper addresses dynamic data-driven signature detection in mobile robots. The core concept of the paper is built upon the principles of Symbolic Dynamic Filtering (SDF) that has been recently reported in literature for extraction of relevant information (i.e., features) in complex dynamical systems. The objective here is to identify the robot behavior in real time as accurately as possible. Two different approaches to classifier design are presented in the paper; the first one is Bayesian and the second is based on measures of formal languages. The proposed methods have been experimentally validated on a networked robotic testbed to detect and identify the type and motion profile of the robots under consideration.

A real time implementable All-Pair Dynamic Planning Algorithm for robot navigation based on the renormalized measure of probabilistic regular languages

The recently reported v planning algorithm is modified to handle on-the-fly dynamic updates to the obstacle map. The modified algorithm called All-Pair-Dynamic-Planning(APDP), models the problem of robot path planning in the framework of finite state probabilistic automata and solves the all-pair planning problem in one setting. We use the concept of renormalized measure of regular languages to plan paths with automated trade-off between path length and robustness under dynamic uncertainties, from any starting location to any goal in the given map. The dynamic updating feature of APDP efficiently updates path plans to incorporate newly learnt information about the working environment.