Path finding problems involve identification of a plan for conflict free movement of agents over a common road network. Most approaches to this problem handle the agents as point objects, wherein the size of the agent is significantly smaller than the road on which it travels. In this paper, we consider spatially extended agents which have a size comparable to the length of the road on which they travel. An optimal multi agent path finding approach for spatially-extended agents was proposed in the eXtended Conflict Based Search (XCBS) algorithm. As XCBS resolves only a pair of conflicts at a time, it results in deeper search trees in case of cascading or multiple (more than two agent) conflicts at a given location. This issue is addressed in eXtended Conflict Based Search with Awareness (XCBS-A) in which an agent uses awareness of other agents' plans to make its own plan. In this paper, we explore XCBS-A in greater detail, we theoretically prove its completeness and empirically demonstrate its performance with other algorithms in terms of variances in road characteristics, agent characteristics and plan characteristics. We demonstrate the distributive nature of the algorithm by evaluating its performance when distributed over multiple machines. XCBS-A generates a huge search space impacting its efficiency in terms of memory; to address this we propose an approach for memory-efficiency and empirically demonstrate the performance of the algorithm. The nature of XCBS-A is such that it may lead to suboptimal solutions, hence the final contribution of this paper is an enhanced approach, XCBS-Local Awareness (XCBS-LA) which we prove will be optimal and complete.

This paper presents use of a cost graph as a representation of a multi-floor building to enable the multi-floor autonomous navigation capability for a team of robot(s). A method for global path planning on this cost graph have been presented. A navigation stack provides a framework for building autonomous navigation capabilities. A navigation stack which enables use of the proposed approach for navigation in a multi-floor building and enables multi-robot operations has been detailed. The improvements provided by the proposed navigation stack over the existing ROS (Robot Operating System) navigation stack have been explained. A way to integrate multiple local path-execution nodes which can combine together to execute the planned global path has also been explained. The paper also demonstrates the reuse of existing ROS compliant source codes for implementation of the proposed navigation stack, thereby optimizing the use of proven and established technology. Further, the extensions to different components of the existing ROS navigation stack, definition of new ROS messages and action definitions, to enable interaction between the components of the stack has been explained. The paper concludes with a brief study on how the proposed stack can be used for multi-robot operations.

Extended Conflict-Based Search (XCBS) is a distributed agent-based approach which has been used for path finding and scheduling of spatially extended agents on a traversable network. The algorithm arrives at an optimal schedule while resolving conflicts between pairs of agents one at a time. In this chapter, we propose XCBS with Awareness wherein a conflicting agent plan is resolved with respect to the proposed route plan of other agents. The approach allows multiple conflicts to be resolved simultaneously, avoids cascading conflicts in the new plans and shows an improved efficiency in terms of nodes explored and time taken to arrive at the solution.

The problem of domain adaptation has been extensively studied for object classification task. However, this problem has not been as well studied for recognizing actions. While, object recognition is well understood, the diverse variety of videos in action recognition make the task of addressing domain shift to be more challenging. We address this problem by proposing a new novel adaptation technique that we term as unsupervised deep action domain adaptation (U-DADA). The main concept that we propose is that of explicitly modeling density based adaptation and using them while adapting domains for recognizing actions. We show that these techniques work well both for domain adaptation through adversarial learning to obtain invariant features or explicitly reducing the domain shift between distributions. The method is shown to work well using existing benchmark datasets such as UCF50, UCF101, HMDB51 and Olympic Sports. As a pioneering effort in the area of deep action adaptation, we are presenting several benchmark results and techniques that could serve as baselines to guide future research in this area.

The objective of this letter is to enhance the accuracy of the monocular visual odometry estimation. The proposed approach estimates pose of the camera via direct approach, by minimizing a novel photometric residue between the current image and warped version of the consecutive image. The novel photometric residue incorporates orientation information accessible from complementary sensor to overcome the inalienable nonlinearity of pose estimation, accordingly lessening the susceptibility to trajectory noise. In addition, this letter introduces a homography-based rotation motion estimation embedded into the photometric residue to reduce tracking failure amid rotations, which could be attributed to lesser number of features with parallax. The proposed methodology permits to integrate Gaussian mixture based prior into visual odometry estimation to cater for changes in motion type, from general to rotation and other way round. Comprehensive experiments have been conducted to exhibit the efficacy of the proposed solution over the best in class visual odometry and visual-inertial systems.

This paper discusses an indoor robotic system that integrates a state-of-the-art object detection algorithm trained with data augmented for an indoor scenario and enabled with mechanisms to localize and position objects in 3D and display them interactively to a user. Size, weight, and power constraints in a mobile robot constrain the type of computing hardware that can be integrated with the robotic platform. However, on the other hand, the robot’s mobility if leveraged properly can provide enough opportunity to detect objects from different distances and viewpoints as the robot approaches them giving more robust results. This work adapts a CNN-based algorithm, YOLO, to run on a GPU-enabled board, the Jetson TX1. An innovative method to calculate the object position in the 3D environment map is discussed along with the problems therein, such as that of duplicate detections that need to be suppressed. Since multiple objects of different or same class may be detected, the user is overloaded with information and management of the visualization through human–machine interaction gains an important role. A scheme for informative display of objects is implemented which lets the user interactively view object images as well as their position in the scene. The complete robotic system including the interactive visualization tool can be put to various uses such as search and rescue, indoor assistance, patrolling and surveillance.

In the main text, both the proposed approaches have been described in detail. The methods have been extensively evaluated using three sets of multi-domain action datasets. In the main text, due to space constraint, only the results for UO (UCF50 and Olympic Sports) and KMS (KTH, MSR Action II and Sonycam) datasets were presented and discussed and the additional results have been included in the supplementary material. We start with the algorithmic details of Action Modeling on Latent Space (AMLS) approach. In the next section, we describe the Symmetrized KL Divergence measure and then discuss the KMS and UO datasets with few example images. In Section 5, we describe our third dataset collection (HU) comprising five common classes of HMDB51 and UCF50 and present its domain adaptation results. In the next section, we discuss the qualitative analysis of the results for the HU dataset and present some of the negative examples observed in our experiments. Finally, in Section 7, we discuss the hyper-parameters and compare some of the results for their different choices.

Although videos appear to be very high-dimensional in terms of duration × frame-rate × resolution, temporal smoothness constraints ensure that the intrinsic dimensionality for videos is much lower. In this paper, we use this idea for investigating Domain Adaptation (DA) in videos, an area that remains under-explored. An approach that has worked well for the image DA is based on the subspace modeling of the source and target domains, which works under the assumption that the two domains share a latent subspace where the domain shift can be reduced or eliminated. In this paper, first we extend three subspace based image DA techniques for human action recognition and then combine it with our proposed Eclectic Domain Mixing (EDM) approach to improve the effectiveness of the DA. Further, we use discrepancy measures such as Symmetrized KL Divergence and Target Density Around Source for empirical study of the proposed EDM approach. While, this work mainly focuses on Domain Adaptation in videos, for completeness of the study, we comprehensively evaluate our approach using both object and action datasets. In this paper, we have achieved consistent improvements over chosen baselines and obtained some state-of-the-art results for the datasets.

This paper proposes a path planning strategy for an Autonomous Ground Vehicle (AGV) navigating in a partially known environment. Global path planning is performed by first using a spatial database of the region to be traversed containing selected attributes such as height data and soil information from a suitable spatial database. The database is processed using a biomimetic swarm algorithm that is inspired by the nest building strategies followed by termites. Local path planning is performed online utilizing information regarding contingencies that affect the safe navigation of the AGV from various sensors. The simulation discussed has been implemented on the open source Player-Stage-Gazebo platform.

Autonomous indoor robotic systems need to perform tasks across multiple floors, such as mapping, search. The floor maps for each floor can be aligned only when the relative position and orientation are known. This can be achieved if the mapping and localization continues as a platform reaches one floor to the next through the staircases. The present work evolves a system which allows a tracked platform navigate through staircases without slippages or tipping, while ensuring that its localization and mapping of the stairway continues. The effort included three main contributions. First, a novel real-time vision-based alignment algorithm that aligns the platform to the staircase is executed before ascending and thereafter, while on-stair, another novel mechanism that continuously keeps the platform aligned with the staircase is executed. For descent, an improved proximity sensor-based algorithm is proposed and analyzed. Some important challenges in mapping and localization are overcome using a sequence of different mapping methods. The staircase mapping algorithm is switched to floor mapping using a novel algorithm for disembark detection using proximity sensor data. The navigation, mapping and localization algorithms work in parallel, allowing the system to perform tasks autonomously across floors. The algorithms developed are independent of slope, width, height, or number of steps, as long as the stairs are straight and regular.

PubMed is a comprehensive database of abstracts and references of a large number of publications in the biomedical domain. Curation of structured connectivity databases creates an easy access point to the wealth of neuroanatomical connectivity information reported in the literature over years. Manual curation of such databases is time consuming and labor intensive. We present a Case Based Reasoning (CBR) approach to automatically compile connectivity status between brain region mentions in text. We focus on the Case Retrieval part of the CBR cycle and present three Instance based learning techniques to retrieve similar cases from the case base. These techniques use varied case representations ranging from surface level features to richer syntax based features. We have experimented with diverse similarity measures and feature weighting schemes for each technique. The three techniques have been evaluated and compared using a benchmark dataset from PubMed and it was found that the one using deep syntactic features gives the best trade off between Precision and Recall. In this study, we have explored issues pertaining to representation of, and retrieval over textual cases. It is envisaged that the ideas presented in the paper can be adapted to needs of other textual CBR domains as well.

Pedestrian detection is a key problem in night vision processing with a dozen of applications that will positively impact the performance of autonomous systems. Despite significant progress, our study shows that performance of state-of-the-art thermal image pedestrian detectors still has much room for improvement. The purpose of this paper is to overcome the challenge faced by the thermal image pedestrian detectors, which employ intensity based Region Of Interest (ROI) extraction followed by feature based validation. The most striking disadvantage faced by the first module, ROI extraction, is the failed detection of cloth insulted parts. To overcome this setback, this paper employs an algorithm and a principle of region growing pursuit tuned to the scale of the pedestrian. The statistics subtended by the pedestrian drastically vary with the scale and deviation from normality approach facilitates scale detection. Further, the paper offers an adaptive mathematical threshold to resolve the problem of subtracting the background while extracting cloth insulated parts as well. The inherent false positives of the ROI extraction module are limited by the choice of good features in pedestrian validation step. One such feature is curvelet feature, which has found its use extensively in optical images, but has as yet no reported results in thermal images. This has been used to arrive at a pedestrian detector with a reduced false positive rate. This work is the first venture made to scrutinize the utility of curvelet for characterizing pedestrians in thermal images. Attempt has also been made to improve the speed of curvelet transform computation. The classification task is realized through the use of the well known methodology of Support Vector Machines (SVMs). The proposed method is substantiated with qualified evaluation methodologies that permits us to carry out probing and informative comparisons across state-of-theart features, including deep learning methods, with six standard and in-house databases. With reference to deep learning, our algorithm exhibits comparable performance. More important is that it has significant lower requirements in terms of compute power and memory, thus making it more relevant for depolyment in resource constrained platforms with significant size, weight and power constraints. (C) 2016 Elsevier B.V. All rights reserved.

This paper presents a survey of Simultaneous Localization And Mapping (SLAM) algorithms for unmanned ground robots. SLAM is the process of creating a map of the environment, sometimes unknown a priori, while at the same time localizing the robot in the same map. The map could be one of different types i.e. metrical, topological, hybrid or semantic. In this paper, the classification of algorithms is done in three classes: (i) Metric map generating approaches, (ii) Qualitative map generating approaches, and (iii) Hybrid map generating approaches. SLAM algorithms for both static and dynamic environments have been surveyed. The algorithms in each class are further divided based on the techniques used. The survey in this paper presents the current state-of-the-art methods, including important landmark works reported in the literature.

The Convoy Movement Problem (CMP) is a common military scenario that requires a clash-free plan for moving large convoys from one location to another with constraints on time and space. This article proposes a distributed agent-based approach for CMP that meets most on-field deployment constraints. The solution is based on a multi-agent path-finding algorithm called conflict-based search (CBS) for point objects. Convoys can't be regarded as point objects: their length can be significantly longer than the edges on which they travel. Hence, a modification to the CBS algorithm makes it applicable to spatially extended objects. This eXtended CBS (XCBS) algorithm has been used to solve the CMP while meeting most of its constraints. The algorithm has been tested on road networks with as many as 40,000 vertices and 79,600 edges for 800 convoys.

Sudoku is a logic-based, combinatorial number placement puzzle, with the objective of filling an n 2 ×n 2 grid of cells with digits so that each column, row and n×n sub-grid (box) that compose the grid contain all the digits from 1 to n 2 . In this paper we study the problem of implementing the Sudoku solving algorithm through the membrane computing paradigm. The devices of this paradigm are called P systems. The solution is obtained by first finding a unique candidate in rows, columns and boxes using the constraints, and then applying various Sudoku solving strategies. Finally, if every strategy fails, the solution is searched through a brute force method. If the solution is obtainable, the P system returns the object YES to the environment. If the Sudoku is unsolvable, the P system returns the object NO to the environment and the system eventually halts. The main contribution of this paper is that the basic P System model defined in [1] has been extended further with additional powerful strategies based on the deductive logic popular with human solvers. The objectives of this effort are twofold (0 to study the very expressive, compact & elegant representation system membrane computing offers (ii) to study the resilience & flexibility offered by the representation to permit addition of new strategies; this being an important requirement in real life applications. The results are presented and discussed for some of the most difficult Sudoku problems available in open source literature. The model will find a solution if it exists. In this sense the model is a generic Sudoku solver.

This paper presents a review of various technologies for autonomous movement of a robot. Path planning is the process of generating a collision free path to the goal. Simultaneous Localization And Mapping (SLAM) is the process of creating a map of the environment while at same time localizing in the same map. Path planning and SLAM are critical for autonomous movement of the robot. This papers discusses different kinds of algorithms for path planning. This paper also describes the methods to incorporate the non-holomic constraints of a robot in the solution. Metrical map generating approaches, qualitative map generating approaches and hybrid map generating approaches for SLAM are also discussed.

A dominant solution concept of non-cooperative game theory is the concept of Nash equilibrium. A Nash equilibrium is a strategy profile from where unilateral deviations do not pay. A nice property of this concept is the well known fact that every finite game has at least one Nash equilibrium. The proof given by Nash (1950) is based on Brouwer fixed point theorem which is very non-constructive. A natural question to ask is whether Nash equilibrium can be computed efficiently. This is still unknown in terms of complexity. Very recently (Daskalakis et al., 2006), it has been shown that the computation of Nash equilibrium is PPAD-complete; which is a new complexity class introduced by Papadimitriou (1994) mainly to capture the complexity of Nash equilibrium. This note tries to give a brief survey of the various algorithms that exist in the literature for computing the Nash equilibrium in finite games. Also, it will briefly touch upon the various existential conditions for Nash equilibrium for infinite games.

Resource allocation in agent societies, which in current literature, is also called as Multi-Agent Resource Allocation (MARA) problem, is the problem of deciding how to distribute a number of resources among a number of agents. It is a growing area of research at the interface of economics and computer science. In this paper, we study the problem of designing a mechanism to allocate a fixed number of objects to agents in a strategy-proof and pareto-efficient way, when each agent has a quota of varying number and indifference of ranking among a set of bundles is allowed. Each agent i needs ai (≥1) many objects from the set of objects R, while having preferences over the set of bundles of size ai. The motivation for us to study such a problem comes from a problem called the site selection problem, which arises often in the military domain. In this paper, we model this site selection problem as a MARA problem. In our model, though we consider a restricted preference, which is justified by the motivating example that is the site selection problem, our algorithm can also accommodate unrestricted preference, that is preference over all possible bundles of size ai. Our main result (Theorem A) ensures that the algorithm developed satisfies both pareto-efficiency and strategy-proofness.

The OODA loop is a common framework in which military decision making is discussed and is an abstraction of the sequence of events that must take place in any military engagement. Of the four components of the OODA loop, the first three involve processing of information, comprising information gathering, information distribution, information analysis, information understanding and deciding how to act upon this information. To achieve higher operational tempo, all components of the OODA loop need to be accelerated. While speed-ups in information gathering and distribution can be attained by well implemented networking, information analysis, understanding and decision making can prove to be severe bottlenecks to the operational tempo. In this context, we discuss the importance of Decision Support Systems (DSS) that dynamically respond to changing scenarios. DSSs developed using Semantic Web techniques and conventional Al-based search mechanisms like A* have the potential to respond to evolving situations providing autonomous capabilities to military applications. This can contribute towards swifter execution of the OODA loop with little human intervention. This paper, discusses in detail, the design and implementation of a DSS for a prototypical problem of convoy scheduling. Convoy Scheduling is a common military operation involving planning a convoy move from one location to another. The problem domain of convoy scheduling has many complexities and various constraints to be satisfied. Several realistically defined scheduling constraints have been incorporated, including permissible convoy speed, movement time, availability of road etc., to name a few. Thus Convoy Scheduling DSS in itself proves to be a self sufficient example to demonstrate the purpose of this paper. Further, the use of ontologies in representing information pertaining to military and geographical domains is demonstrated. The use of A*, a tree-based search algorithm, for optimal clash-free convoy scheduling has also been highlighted. Finally, a framework for a system incorporating this Convoy Scheduling DSS to respond dynamically to various events that may affect scheduled convoy plan is proposed. The responses may include diverting a convoy, inducing forced wait etc. depending on the trade-offs and overall implications.