Search for: All records

Abstract Photoacoustic microscopy (PAM) systems often face challenges in simultaneously achieving high speed, high resolution, high sensitivity, and a large field of view (FOV). To address this challenge, we have developed dual-channel PAM (DC-PAM) that can expand the FOV without compromising the imaging speed, detection sensitivity, or spatial resolution. DC-PAM has two identical, independent channels of laser excitation and acoustic detection. It exploits two facets of a single hexagon scanner to concurrently steer the dual excitation laser beams and the resultant acoustic waves. DC-PAM achieves an ultra-wide FOV of 22.5 × 24 mm² with a total functional imaging time of ~15 s. Proof-of-concept experiments were conducted using DC-PAM on freely-swimming zebrafish, hypoxia-challenged mice, and sleeping glassfrogs, all of which benefit from the large FOV and high imaging speed to track the dynamic and physiological processes at the whole-organ or whole-body level. These applications demonstrate the potential of DC-PAM for a wide range of biological studies. 

Reinforcement Learning (RL) has shown significant promise in automated portfolio management; however, effectively balancing risk and return remains a central challenge, as many models fail to adapt to dynamically changing market conditions. We propose Meta-controlled Agents for a Risk-aware System (MARS), a novel framework addressing this through a multi-agent, risk-aware approach. MARS replaces monolithic models with a Heterogeneous Agent Ensemble, where each agent’s unique risk profile is enforced by a Safety-Critic network to span behaviors from capital preservation to aggressive growth. A high-level Meta-Adaptive Controller (MAC) dynamically orchestrates this ensemble, shifting reliance between conservative and aggressive agents to minimize drawdown during downturns while seizing opportunities in bull markets. This two-tiered structure leverages behavioral diversity rather than explicit feature engineering to ensure a disciplined portfolio robust across market regimes. Experiments on major international indexes confirm that our framework significantly reduces maximum drawdown and volatility while maintaining competitive returns. 

Multigroup discriminant analysis is an important supervised learning technique in the classification framework, with applications in various disciplines. Its objective is to approximate underlying class distributions based on data attributes or features. After the class distributions are estimated, the classification task can be readily carried out for data points with unknown labels. Linear discriminant analysis (LDA) as well as quadratic discriminant analysis (QDA) are statistical procedures widely utilized by practitioners due to their practicality and generally good performance. Both procedures rely on the assumption of normally distributed classes and can be affected by deviations from multivariate normality. To address this model deficiency, we propose an extension of LDA and QDA that relies on the idea of transformation and can readily accommodate asymmetry and skewness in data classes. Through the set of simulation studies and applications to real-life data sets, we demonstrate that the developed technique is promising as it demonstrates superior performance over competitors in a variety of cases. 

Autonomous systems built on ROS2 are increasingly deployed in safety- and performance-critical domains such as autonomous driving and mobile robotics. While existing research has proposed various timing analyses and scheduling strategies for ROS2, many rely on simplified assumptions that do not hold in real-world applications. In this paper, we present a detailed empirical study of ROS2-based autonomous applications, uncovering underexplored runtime behaviors that significantly impact both real-time and functional performance. These include the importance of partial cause-effect chains, dynamic execution paths and timing variability, non-FIFO data access patterns, and computation threads uncontrolled by ROS2 executors. We extend an existing tracing tool to support Transform Library and ROS2’s Action entity, enabling reconstruction and analysis of realistic cause-effect chains. Our findings are validated through experiments in simulated autonomous robot scenarios and a case study using the Autoware autonomous driving framework. Together, our results highlight the need for rethinking ROS2 modeling, scheduling and analysis to better reflect the realities of autonomous systems. 

Deep learning–based classifiers are widely used for perception in autonomous Cyber-Physical Systems (CPS’s). However, such classifiers rarely offer guarantees of perfect accuracy while being optimized for efficiency. To support safety-critical perception, ensembles of multiple different classifiers working in concert are typically used. Since CPS’s interact with the physical world continuously, it is not unreasonable to expect dependencies among successive inputs in a stream of sensor data. Prior work introduced a classification technique that leverages these inter-input dependencies to reduce the average time to successful classification using classifier ensembles. In this paper, we propose generalizations to this classification technique, both in the improved generation of classifier cascades and the modeling of temporal dependencies. We demonstrate, through theoretical analysis and numerical evaluation, that our approach achieves further reductions in average classification latency compared to the prior methods.