Abstract

Quadrotors are becoming more and more essen- tial for applications such as payload delivery, inspection and search-and-rescue. Such operations pose considerable control challenges, especially when various (a priori unbounded) state- dependent unknown dynamics arises from payload variations, aerodynamic effects and from reaction forces while operating close to the ground or in a confined space. However, existing adaptive control strategies for quadrotors cannot handle un- known state-dependent uncertainties. We address such unsolved control challenge in this work via a novel adaptive method for artificial time delay control, where unknown dynamics is robustly compensated by using input and state measurements collected at immediate past time instant (i.e., artificially delayed). Closed-loop stability is established via Lyapunov theory. The effectiveness of this controller is validated using experimental results

Abstract

This work introduces a new single-stage adaptive controller for Euler-Lagrange systems with nonholonomic con- straints. The proposed mechanism provides a simpler design philosophy compared to double-stage mechanisms (that ad- dress kinematics and dynamics in two steps), while achieving analogous stability properties, i.e. stability of both original and internal states. Meanwhile, we do not require direct access to the internal states as required in state-of-the-art single-stage mechanisms. The proposed approach is studied via Lyapunov analysis, validated numerically on wheeled mobile robot dy- namics and compared to a standard double-stage approach

Abstract

The high maneuverability of fixed-wing un- manned aerial vehicles (UAVs) exposes these systems to several dynamical and parametric uncertainties, severely affecting the fidelity of modeling and causing limited guid- ance autonomy. This article shows enhanced autonomy via adaptation mechanisms embedded in the guidance law: a vector-field method is proposed that does not require a priori knowledge of the UAV course time constant, cou- pling effects, and wind amplitude/direction. Stability and performance are assessed using the Lyapunov theory. The method is tested on software-in-the loop and hardware-in- the-loop UAV platforms, showing that the proposed guid- ance law outperforms state-of-the-art guidance controllers and standard vector-field approaches in the presence of significant uncertainty. Index Terms—Adaptive guidance, adaptive sliding-mode control, fixed-wing unmanned aerial vehicles (UAV), un- known dynamics, vector field (VF).

Abstract

In this work, we describe an end-to-end deep learning architecture for Heart Murmur Detection from Phonocardiogram(PCG) recordings as part of The George B. Moody PhysioNet Challenge 2022. Our team, “Team IIITH” received a weighted accuracy score of 0.708 (ranked 19th out of 40 teams) and Challenge cost of 13264 (ranked 22nd out of 39 teams) on the official hidden test set. In our approach, the PCG recordings are first downsampled to 1000 Hz before being passed through a Butterworth’s low and high pass filter to remove baseline wanders and high-frequency noise present in the recordings. The PCG recordings are then broken down into 10-second segments and normalized to bring all trainable samples to the same size. To extract embeddings more efficiently, we built a custom 1-dimensional Residual Network (ResNet) where the 10-second inputs are passed through variable-sized kernel ResNets in parallel, before being concatenated and passed through the next ResNet layer to account for different length dependencies across the PCG signal. The output of this custom ResNet is then fed to a 2-layer feed-forward network for final classification. Cross-Entropy Loss with class weights was employed to account for class imbalance. Our approach obtained a 5-fold Cross-Validation weighted accuracy score of 0.71 and challenge cost score of 12067 on the training set.

Abstract

Neuroimaging techniques such as fMRI record brain activation while participants experience a stimulus. The concreteness of concepts defines how well our brain is able to imagine them. We hypothesise that brain activation would be distinctly different when participants view stimuli corresponding to concrete words versus abstract words. Specifically, we expect primary sensory areas to engage more in the former. To study this, we used a multi-view dataset with each concept being displayed as a picture, as a word in a sentence and also as a wordcloud [1]. We investigate how well a machine learning-based decoder trained on concrete concepts can decode both concrete and abstract concepts and vice versa for the model trained on abstract concepts. We find that a model trained on stimuli related to concrete concepts yields better accuracy than the model trained on abstract concepts. We also explore the contribution of voxels from different brain regions in this decoding process. Analysis of the contribution of different brain networks reveals exciting cognitive insights.

Abstract

We identify a large family of abelian codes that achieve the capacity of the binary erasure channel (BEC) under bit-MAP decoding. The codes in this family have rich automorphism groups, their lengths are odd integers, and they can asymptotically (in the block length) achieve any code rate. This family contains codes of prime power block lengths that were originally identified by Berman (1967) and later inves- tigated by Blackmore and Norton (2001), and also contains their generalization to any odd block length. We use Rajan and Siddiqi’s (1992) characterization of abelian codes using discrete Fourier transform (DFT) to identify our code family and study their automorphism groups. We then use a result of Kumar, Calderbank and Pfister (2016) that relates the automorphism group of a code to its performance in the BEC to show that this code family achieves BEC capacity. The full version of this paper, including the proofs of all claims and simulation results, is available online

Abstract

We identify a family of binary codes whose structure is similar to Reed-Muller (RM) codes and which include RM codes as a strict subclass. The codes in this family are denoted as Cn(r, m), and their duals are denoted as Bn(r, m). The length of these codes is n m, where n ≥ 2, and r is their ‘order’. When n = 2, Cn(r, m) is the RM code of order r and length 2 m. The special case of these codes corresponding to n being an odd prime was studied by Berman (1967) and Blackmore and Norton (2001). Following the terminology introduced by Blackmore and Norton, we refer to Bn(r, m) as the Berman code and Cn(r, m) as the dual Berman code. We identify these codes using a recursive Plotkin-like construction, and we show that these codes have a rich automorphism group, they are generated by the minimum weight codewords, and that they can be decoded up to half the minimum distance efficiently. Using a result of Kumar et al. (2016), we show that these codes achieve the capacity of the binary erasure channel (BEC) under bit-MAP decoding. Furthermore, except double transitivity, they satisfy all the code properties used by Reeves and Pfister to show that RM codes achieve the capacity of binary-input memoryless symmetric channels. Finally, when n is odd, we identify a large class of abelian codes that includes Bn(r, m) and Cn(r, m) and which achieves BEC capacity

Abstract

We identify a family of binary codes whose structure is similar to Reed-Muller (RM) codes and which include RM codes as a strict subclass. The codes in this family are denoted as Cn(r,m), and their duals are denoted as Bn(r,m). The length of these codes is nm, where n≥2, and r is their `order'. When n=2, Cn(r,m) is the RM code of order r and length 2m. The special case of these codes corresponding to n being an odd prime was studied by Berman (1967) and Blackmore and Norton (2001). Following the terminology introduced by Blackmore and Norton, we refer to Bn(r,m) as the Berman code and Cn(r,m) as the dual Berman code. We identify these codes using a recursive Plotkin-like construction, and we show that these codes have a rich automorphism group, they are generated by the minimum weight codewords, and that they can be decoded up to half the minimum distance efficiently. Using a result of Kumar et al. (2016), we show that these codes achieve the capacity of the binary erasure channel (BEC) under bit-MAP decoding. Furthermore, except double transitivity, they satisfy all the code properties used by Reeves and Pfister to show that RM codes achieve the capacity of binary-input memoryless symmetric channels. Finally, when n is odd, we identify a large class of abelian codes that includes Bn(r,m) and Cn(r,m) and which achieves BEC capacity.

Abstract

Environmental flow (Eflow) allocations of reservoir-river systems do not adequately address the connections between reservoir inflows, releases, and corresponding downstream river water quality rendering water resources management difficult. The study integrated the Soil and Water Assessment Tool (SWAT) for reservoir inflows estimation and corresponding releases with Global Environmental Flow Calculator (GEFC) for Eflows allocations to simulate the resulting Dissolved Oxygen (DO) of the downstream river stretch using QUAL2K. The study considered various plausible scenarios of inflows (10 to 20% reduction), pollution scenarios of Biological Oxygen Demand (BOD) (0 to 100% treatment) for arriving drains by altering headwaters DO (4 to 8 mg/l). Eflow water quality charts were developed for the regulation of reservoir downstream river water quality for Bhadra river in India as case study. The study revealed that by maintaining the headwater DO in conjunction with BOD treatment of drains and with sufficient Eflow allocation, downstream river water quality can be improved. By maintaining a headwater DO of 7 mg/l with BOD treatment of drains of 25% and by providing Eflow allocation of 40.12 m3/s with class D, the Bhadravathi river stretch has shown an improved river water quality with about 6.65 mg/l of average DO.

Abstract

This paper presents a simple and effective framework to combine various data-driven machine learning (ML) algorithms for short-range reservoir inflow forecasting, including the large-scale climate phenomenon indices addressing forecasting uncertainty. Random Forest (RF), Gradient Boosting Regressor (GBR), K-Nearest Neighbors Regressor (KNN), and Long Short-Term Memory (LSTM) were employed for predicting daily reservoir inflows considering various climate phenomenon indices (e.g., Arctic Oscillation, North Atlantic Oscillation, and Southern Oscillation Index) and hydroclimatic variables (e.g., precipitation), accounting for time-lag effects. After training the individual ML algorithm, a framework was developed to create an ensemble model using a robust weighted voting regressor (VR) method to quantify forecasting uncertainty and to improve model performances. The results of the study reveal that, for 2-day forecasts, the LSTM approach has the greatest influence on prediction accuracy, followed comparably by each model. However, none of the four models seem to be noticeably superior to the VR method, regardless of the prediction lead time. The developed framework was examined on a tropical reservoir, Bhadra reservoir Tunga-Bhadra River, located in India