Sim2Real (Simulation to Reality) techniques have gained prominence in robotic manipulation and motion planning due to their ability to enhance success rates by enabling agents to test and evaluate various policies and trajectories. In this paper, we investigate the advantages of integrating Sim2Real into robotic frameworks. We introduce the Triple Regression Sim2Real framework, which constructs a real-time digital twin. This twin serves as a replica of reality to simulate and evaluate multiple plans before their execution in real-world scenarios. Our triple regression approach addresses the reality gap by: (1) mitigating projection errors between real and simulated camera perspectives through the first two regression models, and (2) detecting discrepancies in robot control using the third regression model. Experiments on 6-DoF grasp and manipulation tasks (where the gripper can approach from any direction) highlight the effectiveness of our framework. Remarkably, with only RGB input images, our method achieves state-of-the-art success rates. This research advances efficient robot training methods and sets the stage for rapid advancements in robotics and automation.
We introduce a new approach to address the task allocation problem in a system of heterogeneous robots comprising of Unmanned Ground Vehicles (UGVs) and Unmanned Aerial Vehicles (UAVs). The proposed model, \texttt{\method}, or \textbf{G}raph \textbf{A}ttention \textbf{T}ask \textbf{A}llocato\textbf{R} aggregates information from neighbors in the multi-robot system, with the aim of achieving joint optimality in the target localization efficiency.Being decentralized, our method is highly robust and adaptable to situations where collaborators may change over time, ensuring the continuity of the mission. We also proposed heterogeneity-aware preprocessing to let all the different types of robots collaborate with a uniform model.The experimental results demonstrate the effectiveness and scalability of the proposed approach in a range of simulated scenarios. The model can allocate targets' positions close to the expert algorithm's result, with a median spatial gap less than a unit length. This approach can be used in multi-robot systems deployed in search and rescue missions, environmental monitoring, and disaster response.
This paper addresses the design of a partly-parallel cascaded FFT-IFFT architecture that does not require any intermediate buffer. Folding can be used to design partly-parallel architectures for FFT and IFFT. While many cascaded FFT-IFFT architectures can be designed using various folding sets for the FFT and the IFFT, for a specified folded FFT architecture, there exists a unique folding set to design the IFFT architecture that does not require an intermediate buffer. Such a folding set is designed by processing the output of the FFT as soon as possible (ASAP) in the folded IFFT. Elimination of the intermediate buffer reduces latency and saves area. The proposed approach is also extended to interleaved processing of multi-channel time-series. The proposed FFT-IFFT cascade architecture saves about N/2 memory elements and N/4 clock cycles of latency compared to a design with identical folding sets. For the 2-interleaved FFT-IFFT cascade, the memory and latency savings are, respectively, N/2 units and N/2 clock cycles, compared to a design with identical folding sets.
We present ExBluRF, a novel view synthesis method for extreme motion blurred images based on efficient radiance fields optimization. Our approach consists of two main components: 6-DOF camera trajectory-based motion blur formulation and voxel-based radiance fields. From extremely blurred images, we optimize the sharp radiance fields by jointly estimating the camera trajectories that generate the blurry images. In training, multiple rays along the camera trajectory are accumulated to reconstruct single blurry color, which is equivalent to the physical motion blur operation. We minimize the photo-consistency loss on blurred image space and obtain the sharp radiance fields with camera trajectories that explain the blur of all images. The joint optimization on the blurred image space demands painfully increasing computation and resources proportional to the blur size. Our method solves this problem by replacing the MLP-based framework to low-dimensional 6-DOF camera poses and voxel-based radiance fields. Compared with the existing works, our approach restores much sharper 3D scenes from challenging motion blurred views with the order of 10 times less training time and GPU memory consumption.
In recent years, a certain type of problems have become of interest where one wants to query a trained classifier. Specifically, one wants to find the closest instance to a given input instance such that the classifier's predicted label is changed in a desired way. Examples of these ``inverse classification'' problems are counterfactual explanations, adversarial examples and model inversion. All of them are fundamentally optimization problems over the input instance vector involving a fixed classifier, and it is of interest to achieve a fast solution for interactive or real-time applications. We focus on solving this problem efficiently for two of the most widely used classifiers: logistic regression and softmax classifiers. Owing to special properties of these models, we show that the optimization can be solved in closed form for logistic regression, and iteratively but extremely fast for the softmax classifier. This allows us to solve either case exactly (to nearly machine precision) in a runtime of milliseconds to around a second even for very high-dimensional instances and many classes.
Mass casualty incidents (MCIs) pose a formidable challenge to emergency medical services by overwhelming available resources and personnel. Effective victim assessment is paramount to minimizing casualties during such a crisis. In this paper, we introduce ARTEMIS, an AI-driven Robotic Triage Labeling and Emergency Medical Information System. This system comprises a deep learning model for acuity labeling that is integrated with a robot, that performs the preliminary assessment of injury severity in patients and assigns appropriate triage labels. Additionally, we have developed a frontend (graphical user interface) that is updated by the robots in real time and is accessible to the first responders. To validate the reliability of our proposed algorithmic triage protocol, we employed an off-the-shelf robot kit equipped with sensors for vital sign acquisition. A controlled laboratory simulation of an MCI was conducted to assess the system's performance and effectiveness in real-world scenarios resulting in a triage-level classification accuracy of 92%. This noteworthy achievement underscores the model's proficiency in discerning crucial patterns for accurate triage classification, showcasing its promising potential in healthcare applications.
Native Language Identification (NLI) intends to classify an author's native language based on their writing in another language. Historically, the task has heavily relied on time-consuming linguistic feature engineering, and transformer-based NLI models have thus far failed to offer effective, practical alternatives. The current work investigates if input size is a limiting factor, and shows that classifiers trained using Big Bird embeddings outperform linguistic feature engineering models by a large margin on the Reddit-L2 dataset. Additionally, we provide further insight into input length dependencies, show consistent out-of-sample performance, and qualitatively analyze the embedding space. Given the effectiveness and computational efficiency of this method, we believe it offers a promising avenue for future NLI work.
In the area of learning-driven artificial intelligence advancement, the integration of machine learning (ML) into self-driving (SD) technology stands as an impressive engineering feat. Yet, in real-world applications outside the confines of controlled laboratory scenarios, the deployment of self-driving technology assumes a life-critical role, necessitating heightened attention from researchers towards both safety and efficiency. To illustrate, when a self-driving model encounters an unfamiliar environment in real-time execution, the focus must not solely revolve around enhancing its anticipated performance; equal consideration must be given to ensuring its execution or real-time adaptation maintains a requisite level of safety. This study introduces an algorithm for online meta-reinforcement learning, employing lookahead symbolic constraints based on \emph{Neurosymbolic Meta-Reinforcement Lookahead Learning} (NUMERLA). NUMERLA proposes a lookahead updating mechanism that harmonizes the efficiency of online adaptations with the overarching goal of ensuring long-term safety. Experimental results demonstrate NUMERLA confers the self-driving agent with the capacity for real-time adaptability, leading to safe and self-adaptive driving under non-stationary urban human-vehicle interaction scenarios.
This white paper aims to briefly describe a proposed article that will provide a thorough comparative study of waveforms designed to exploit the features of doubly-dispersive channels arising in heterogeneous high-mobility scenarios as expected in the beyond fifth generation (B5G) and sixth generation (6G), in relation to their suitability to integrated sensing and communications (ISAC) systems. In particular, the full article will compare the well-established delay-Doppler domain-based orthognal time frequency space (OTFS) and the recently proposed chirp domain-based affine frequency division multiplexing (AFDM) waveforms. Both these waveforms are designed based on a full delay- Doppler representation of the time variant (TV) multipath channel, yielding not only robustness and orthogonality of information symbols in high-mobility scenarios, but also a beneficial implication for environment target detection through the inherent capability of estimating the path delay and Doppler shifts, which are standard radar parameters. These modulation schemes are distinct candidates for ISAC in B5G/6G systems, such that a thorough study of their advantages, shortcomings, implications to signal processing, and performance of communication and sensing functions are well in order. In light of the above, a sample of the intended contribution (Special Issue paper) is provided below.
Evaluating open-domain dialogue systems is challenging for reasons such as the one-to-many problem, i.e., many appropriate responses other than just the golden response. As of now, automatic evaluation methods need better consistency with humans, while reliable human evaluation can be time- and cost-intensive. To this end, we propose the Reference-Assisted Dialogue Evaluation (RADE) approach under the multi-task learning framework, which leverages the pre-created utterance as reference other than the gold response to relief the one-to-many problem. Specifically, RADE explicitly compares reference and the candidate response to predict their overall scores. Moreover, an auxiliary response generation task enhances prediction via a shared encoder. To support RADE, we extend three datasets with additional rated responses other than just a golden response by human annotation. Experiments on our three datasets and two existing benchmarks demonstrate the effectiveness of our method, where Pearson, Spearman, and Kendall correlations with human evaluation outperform state-of-the-art baselines.