Abstract

Existing deep learning based visual servoing approaches regress the relative camera pose between a pair of images. Therefore, they require a huge amount of training data and sometimes fine-tuning for adaptation to a novel scene. Furthermore, current approaches do not consider underlying geometry of the scene and rely on direct estimation of camera pose. Thus, inaccuracies in prediction of the camera pose, especially for distant goals, lead to a degradation in the servoing performance. In this paper, we propose a two-fold solution: (i) We consider optical flow as our visual features, which are predicted using a deep neural network. (ii) These flow features are then systematically integrated with depth estimates provided by another neural network using interaction matrix. We further present an extensive benchmark in a photo-realistic 3D simulation across diverse scenes to study the convergence and generalisation of visual servoing approaches. We show convergence for over 3m and 40 degrees while maintaining precise positioning of under 2cm and 1 degree on our challenging benchmark where the existing approaches that are unable to converge for majority of scenarios for over 1.5m and 20 degrees. Furthermore, we also evaluate our approach for a real scenario on an aerial robot. Our approach generalizes to novel scenarios producing precise and robust servoing performance for 6 degrees of freedom positioning tasks with even large camera transformations without any retraining or fine-tuning.

Abstract

Autonomous cars and fixed-wing aerial vehicles have the so-called non-holonomic kinematics which non-linearly maps control input to states. As a result, trajectory optimization with such a motion model becomes highly non-linear and nonconvex. In this paper, we improve the computational tractability of non-holonomic trajectory optimization by reformulating it in terms of a set of bi-convex cost and constraint functions along with a non-linear penalty. The bi-convex part acts as a relaxation for the non-holonomic trajectory optimization while the residual of the penalty dictates how well its output obeys the non-holonomic behavior. We adopt an alternating minimization approach for solving the reformulated problem and show that it naturally leads to the replacement of the challenging nonlinear penalty with a globally valid convex surrogate. Along with the common cost functions modeling goal-reaching, trajectory smoothness, etc., the proposed optimizer can also accommodate a class of non-linear costs for modeling goal-sets, while retaining the bi-convex structure. We benchmark the proposed optimizer against off-the-shelf solvers implementing sequential quadratic programming and interior-point methods and show that it produces solutions with similar or better cost as the former while significantly outperforming the latter. Furthermore, as compared to both off-the-shelf solvers, the proposed optimizer achieves more than 20x reduction in computation time.

Abstract

In this paper, we present SROM, a novel realtime Simultaneous Localization and Mapping (SLAM) system for autonomous vehicles. The keynote of the paper showcases SROM’s ability to maintain localization at low sampling rates or at high linear or angular velocities where most popular LiDAR based localization approaches get degraded fast. We also demonstrate SROM to be computationally efficient and capable of handling high-speed maneuvers. It also achieves low drifts without the need for any other sensors like IMU and/or GPS.Our method has a two-layer structure wherein first, an approximate estimate of the rotation angle and translation parameters are calculated using a Phase Only Correlation (POC) method.Next, we use this estimate as an initialization for a point-toplane ICP algorithm to obtain fine matching and registration. Another key feature of the proposed algorithm is the removal of dynamic objects before matching the scans. This improves the performance of our system as the dynamic objects can corrupt the matching scheme and derail localization. Our SLAM system can build reliable maps at the same time generating high-quality odometry. We exhaustively evaluated the proposed method in many challenging highways/country/urban sequences from the KITTI dataset and the results demonstrate better accuracy in comparisons to other state-of-the-art methods with reduced computational expense aiding in real-time realizations. We have also integrated our SROM system with our in-house autonomous vehicle and compared it with the state-of the-art methods like LOAM and LeGO-LOAM.

Abstract

Adaptive Sliding Mode Control (ASMC) aims to adapt the switching gain in such a way to cope with possibly unknown uncertainty. In state-of-the-art ASMC methods, a priori boundedness of the uncertainty is crucial to ensure boundedness for the switching gain and uniformly ultimately boundedness. A priori bounded uncertainty might impose a priori bounds on the system state before obtaining closed-loop stability. A design removing this assumption is still missing in literature. A positive answer to this quest is given by this note where a novel ASMC methodology is proposed which does not require a priori bounded uncertainty. An illustrative example is presented to highlight the main features of the approach, after which a general class of Euler–Lagrange systems is taken as a case study to show the applicability of the proposed design.

Abstract

This brief proposes a new adaptive-robust formulation for time-delay control (TDC) under a less-restrictive stability condition. TDC relies on estimating the unknown system dynamics via the artificial introduction of a time delay, often referred to as time-delay estimation (TDE). In conventional TDC, the estimation error, called TDE error, is taken to be upper bounded by a constant under the assumption of small time delay and, most importantly, of a priori bounded states. We highlight the issues of such a conventional methodology via an unstable the upper bound of the TDE error is formulated, which has an explicit dependence on system states and is valid for any chosen time delay. This insight leads to a new TDC design, namely, time-delayed adaptive-robust control (TDARC). The effectiveness of TDARC is substantiated via a multiple-degrees-of-freedom robot.

Abstract

Available control methods for underactuated Euler–Lagrange (EL) systems rely on structure-specific constraints that may be appropriate for some systems, but restrictive for others. A generalized (structure-independent) control framework is to a large extent missing, especially in the presence of uncertainty. This paper introduces an adaptive-robust control framework for a quite general class of uncertain underactuated EL systems. Compared to existing literature, the important attributes of the proposed solution are: (i) avoiding structure-specific restrictions, namely, symmetry condition property of the mass matrix, and a priori bounds on non-actuated states or state derivatives; (ii) considering Coriolis, centripetal, friction and gravity terms to be unknown, while only requiring the knowledge of maximum perturbation around a nominal value of the mass matrix; (iii) handling state-dependent uncertainties irrespective of their linear or nonlinear in parameters structure. These features significantly widen the range of underactuated EL systems the proposed solution can handle in comparison to the available methods. Stability is studied analytically and the performance is verified in simulation using offshore boom crane dynamics.

Abstract

The literature has proven that attaining good transient behavior in leakage-based robust adaptive control of uncertain switched systems is intrinsically challenging. In fact, because the gains of the inactive subsystems must exponentially vanish during inactive times as an effect of leakage action, new learning transients will repeatedly arise at each switching instant. In this paper, a new leakage-based mechanism is designed for robust adaptive control of uncertain switched systems: in contrast to the available designs, the key innovation of the proposed one is that the adaptive gains of the inactive subsystems can be kept constant to their switched-off values, thus preventing vanishing gains. Bounded stability of the closed-loop switched system is guaranteed thanks to the introduction of an auxiliary gain playing the role of leakage. A benchmark example commonly adopted in adaptive switched literature shows that the proposed strategy can consistently improve the transient behavior under various families of switching signals.

Abstract

Frequent pattern mining is one among the popular data mining techniques. Frequent pattern mining approaches extract interesting associations among the items in a given transactional database. The items of the transactional database can be organized as a concept hierarchy. Notably, frequent pattern mining does not distinguish the patterns by analyzing the categories of the items in a given concept hierarchy. In several applications, it is often useful to distinguish among the frequent patterns by analyzing how the items of the pattern are mapped to different categories of the concept hierarchy. In this paper, we propose a new interestingness measure, designated as diversity rank (drank), for capturing the diversity of a given pattern by analyzing the extent to which the items of the pattern are associated with the categories of the corresponding concept hierarchy. Given a transactional database over a set I of items and the corresponding concept hierarchy on I, we propose a methodology to compute the drank of the given pattern. Furthermore, by extending the notion of drank, we propose an approach to improve the diversity and accuracy of association rule-based recommender system. The results of our performance evaluation on the real-world MovieLens dataset demonstrate that the proposed diversity model extracts different kinds of patterns as compared to frequent patterns. Furthermore, our proposed recommender system approach improves the diver sit y performance w.r.t. the existing association rule-based recommender system without significantly compromising the accuracy. Overall, the proposed concept hierarchy-based diverse pattern model provides a scope to develop new approaches for improving the performance of frequent pattern mining-based applications.

Abstract

Due to advances in technology for better output/production, there has been an increased demand for technology with optimized resource utilization in order to conserve resources. Advancements in sensors have improved resource optimization in many fields such as electric power consumption, water management and quality management. In this regard, air conditioner (AC) management is an important aspect of electric power consumption. In this paper, we address the issue of reducing power consumption for air conditioning of a given room. In previous works, the whole room is maintained at a fixed required temperature by inherently assuming that all of the users prefer the same temperature; this is generally not true in practice. In this paper, we propose a sensor-based framework to place and manage ACs for maintaining diverse temperature zones in a given layout to reduce power consumption. Through a simulation study, we demonstrate that the proposed framework indeed has the potential to reduce power consumption significantly as compared to the naive approach by achieving user satisfaction.

Abstract

Pattern mining is an important task of data mining and involves the extraction of interesting associations from large databases. However, developing fast and efficient parallel algorithms for handling large volumes of data is a challenging task. The MapReduce framework enables the distributed processing of huge amounts of data in large-scale distributed environment with robust fault-tolerance. In this paper, we propose a parallel algorithm for extracting coverage patterns. The results of our performance evaluation with real-world and synthetic datasets demonstrate that it is indeed feasible to extract coverage patterns effectively under the MapReduce framework.