Abstract

We focus on the problem of LiDAR point cloud based loop detection (or Finding) and closure (LDC) for mobile robots. State-of-the-art (SOTA) methods directly generate learned embeddings from a given point cloud, require large data augmentation, and are not robust to wide viewpoint variations in 6 Degrees-of-Freedom (DOF). Moreover, the absence of strong priors in an unstructured point cloud leads to highly inaccurate LDC. In this original approach, we propose independent roll and pitch canonicalization of point clouds using a common dominant ground plane. We discretize the canonicalized point clouds along the axis perpendicular to the ground plane leads to images similar to digital elevation maps (DEMs), which expose strong spatial priors in the scene. Our experiments show that LDC based on learnt embeddings from such DEMs is not only data efficient but also significantly more robust, and generalizable than the current SOTA. We report an (average precision for loop detection, mean absolute translation/rotation error) improvement of (8.4, 16.7/5.43)% on the KITTI08 sequence, and (11.0, 34.0/25.4)% on GPR10 sequence, over the current SOTA. To further test the robustness of our technique on point clouds in 6-DOF motion we create and opensource a custom dataset called LidarUrbanFly Dataset (LUF) which consists of point clouds obtained from a LiDAR mounted on a quadrotor. More details on our website https://gsc2001.github.io/FinderNet/

Abstract

Worldover, seismic design of buildings typically follows a prescriptive approach in which designers conform to a series of prescriptive code requirements in terms of both analysis and design procedures. Even though this prescriptive seismic design approach is time-tested and easily understandable by structural designers. In the recent past, performance-based seismic design has started to gain traction among structural designers. The performance-based seismic design allows designers to set performance objectives and design buildings to meet the targeted performance criteria. Due to its flexible nature, performance-based design has proven extremely useful for critical and lifeline buildings like hospitals and tall buildings. With a focus placed on performance objectives, designers utilizing performance-based seismic design are proficient in designing code exceeding buildings efficiently. Despite these cited benefits, performance-based design is still considered an uncommon practice in structural design, particularly in the developing countries. Hence, the present study aims to provide an overview and framework to practice performance-based seismic design. This work identifies and discusses the key differences between the prescriptive- and performance-based seismic design methods and also addresses the significance, application and implementation of performance-based seismic design of buildings. This paper makes an original contribution to the literature through a critical review of how the performance-based design withholds the opportunity to elevate the role of the structural engineers to which they are informed members of the community, where the structures they create not only perform according to design prescriptions, but also perform according to the needs of the owners, engineers, and society.

Abstract

We consider a distributed multi-user secret sharing setting consisting of a dealer, n storage nodes and m secrets where each user demands a t-subset of m secrets. The user downloads shares from the storage nodes based on the designed access structure and reconstructs its secrets. Earlier work in this setting dealt with the case of t = 1; in this work, we generalize it to natural numbers. We identify a necessary condition on the access structures to ensure weak secrecy. We also make a connection between the access structures for this problem and t-disjunct matrices. We apply various t-disjunct matrix constructions in this setting and compare their performance in terms of the number of storage nodes and communication complexity. We also derive bounds on the optimal communication complexity of a distributed secret sharing protocol. Finally, we characterize the capacity region of the DMUSS problem when the access structure is specified

Abstract

Speech is produced by exciting time-varying vocal tract with time-varying impulse-like excitations called epochs. In the literature, epoch extraction methods performed well on clean speech, but, detecting epoch locations from the band-limited signal like telephonic speech is difficult due to loss of information at low frequencies. This paper proposes a Stockwell transform (S-Transform)-based method that can find epochs accurately from the telephonic speech. The frequency-dependent Gaussian window and localization capabilities of S-Transform will reduce the effect of the bandpass nature of the telephonic channel. The telephonic channel is simulated using a 300–3400 Hz bandpass filter. The proposed method is evaluated on five speakers data, namely BDL, SLT, JMK, KED, and RAB, from CMU arctic database. The results are compared with the state-of-the-art methods for both clean speech and telephonic speech. The proposed method produced comparable results with existing methods on clean speech but has shown an improvement of 4.68over state-of-the-art methods.

Abstract

Urbanization, driven by technological advancements, has brought about improved connectivity and efficiency, especially with the rise of Internet of Things (IoT) devices. Smart cities use these innovations to manage resources better and enhance resident’s quality of life. However, implementing smart city initiatives comes with challenges like monitoring, maintaining, and testing urban infrastructure. Digital Twin (DT) entails the connection of physical facilities or devices with their digital counterparts, facilitating real-time monitoring, manipulation, and predictive analysis of their behavior. This concept offers a virtual replica of assets, processes, and systems, enabling insights into their real-time performance and predictive behaviors. By simulating real-world scenarios, DT aids in planning maintenance activities and conducting comprehensive testing, thereby enhancing the resilience and efficiency of smart city systems. Particularly in the context of managing water networks, DT technology holds significant promise. Visualization capabilities provide intuitive insights into the system’s behavior, facilitating informed decision-making. This visualization, coupled with actuation capabilities, enables control actions based on predictive analytics and optimization algorithms, allowing for proactive management of water resources and infrastructure. To this end, in this paper, we present the architecture of WaterTwin, a DT developed for water quality networks in smart city systems. We demonstrate our approach through the use of a water quality network at the smart city living lab, IIIT Hyderabad campus.

Abstract

Risk assessment is vital for comprehending and managing river water quality control. By ensuring a systematic approach for evaluating potential risks, it encourages the adoption of efficient management strategies, the use of informed decision-making, and maintaining of river ecosystems. An effective framework is developed by integrating the empirical based approach with Gumbel-Copula function to estimate the river water quality risk under high water temperatures and low flow rates. Based on Akaike Information Criterion (AIC) among Copula based Gumbel, Gaussian, t, and Clay-ton functions, the Gumbel-Copula function was utilized in the study. The percentiles ranging from 5 to 95 with 5 percentile interval of Stream Flows (SF) and reverse order percentiles of Water Temperatures (WT) are employed in the estimation of risk of river water quality. The current risk quantification approach was implemented using observed datasets on Yamuna (2000 to 2015) and Bhadra (2006 to 2017) river systems. The results reveal that at 50 percentile (SF=380 m3/s and WT=27.38 ℃) the join risk is about 75.29% for Yamuna River and similarly for Bhadra river the risk is about 77.62% (SF=7.20 m3/s and WT=25 ℃).

Abstract

Gauging the knowledge of Pretrained Language Models (PLMs) about facts in niche domains is an important step towards making them better in those domains. In this paper, we aim at evaluating multiple PLMs for their knowledge about world Geography. We contribute (i) a sufficiently sized dataset of masked Geography sentences to probe PLMs on masked token prediction and generation tasks, (ii) benchmark the performance of multiple PLMs on the dataset. We also provide a detailed analysis of the performance of the PLMs on different Geography facts.

Abstract

The emergence of smart grid technology has opened the door for wide-scale automation in decision-making. A distribution company, an integral part of a smart grid system, has to procure electricity from the wholesale market and then sell it to customers in the retail market by publishing attractive tariff contracts. It can deploy autonomous agents to make decisions on its behalf. In this work, we describe the tariff contracts generation strategy of one such autonomous agent, which is based on a Contextual Multi-armed Bandit (ConMAB) based learning technique to generate tariff contracts for various types of customers in the retail market of smart grids. We particularly utilize the Exponential-weight algorithm for Exploration and Exploitation (EXP-3) for ConMAB-based learning. We call our proposed strategy GenerateTariffs-EXP3. Our previous work shows that maintaining an appropriate market share in the retail market yields high net revenue. Thus, we first present a game-theoretic analysis that determines an optimal level of market share. Then we train our proposed strategy to achieve and maintain the suggested level of market share by adapting to the market situation and revising the tariff contracts periodically. We validate our proposed strategy in PowerTAC, a close-to real-world smart grid simulator, and showcase that it is able to maintain the suggested market share.

Abstract

The Multi-Object Navigation (MultiON) task requires a robot to localize an instance (each) of multiple object classes. It is a fundamental task for an assistive robot in a home or a factory. Existing methods for MultiON have viewed this as a direct extension of Object Navigation (ON), the task of localising an instance of one object class, and are pre-sequenced, i.e., the sequence in which the object classes are to be explored is provided in advance. This is a strong limitation in practical applications characterized by dynamic changes. This paper describes a deep reinforcement learning framework for sequence-agnostic MultiON based on an actor-critic architecture and a suitable reward specification. Our framework leverages past experiences and seeks to reward progress toward individual as well as multiple target object classes. We use photo-realistic scenes from the Gibson benchmark dataset in the AI Habitat 3D simulation environment to experimentally show that our method performs better than a pre-sequenced approach and a state of the art ON method extended to MultiON.

Abstract

Speech systems are sensitive to accent variations. This is a challenge in India, which has numerous languages but few linguistic studies on pronunciation variation. The growing number of L2 English speakers reinforces the need to study accents and L1-L2 interactions. We investigate Indian English (IE) accents and report our observations on regional and shared features. Specifically, we observe phonemic variations and phonotactics in speakers’ native languages and apply this to their English pronunciations. We demonstrate the influence of 18 Indian languages on IE by comparing native language features with IE pronunciations obtained from literature studies and phonetically annotated speech. Hence, we validate Indian language influences on IE by justifying pronunciation rules from the perspective of Indian language phonology. We obtain a comprehensive description of generalised and region-specific IE characteristics, which facilitates accent adaptation of existing speech systems.