arXiv subjects
Publications and source records attributed to Rahul Bhadani.
We investigate generalized Kennedy receivers for coherent-state BPSK communication over turbulent free-space optical channels. An ergodic mutual-information framework is developed to jointly optimize the receiver displacement and input prior probabilities under Gamma--Gamma fading.
This paper studies neural ordinary differential equations (neural ODEs) from a control-theoretic perspective using controllability and observability concepts. The neural ODE is represented in a control-affine form to facilitate analysis using tools from nonlinear and linear time-varying (LTV) systems. Controllability is examined through trajectory linearization, where the LTV controllability Gramian provides a local, first-order measure of input influence along a nominal trajectory. Observability is analyzed through output linearization, where the LTV observability Gramian characterizes the local ability to reconstruct system states from output measurements. Koopman-based lifting is considered to extend the analysis to a higher-dimensional representation, and its limitations under multiple equilibria and basin-dependent behavior are discussed. The proposed framework is illustrated on a series RLC circuit. The learned neural ODE reproduces system trajectories and generalizes to unseen initial conditions. The computed Gramians are numerically full rank along the tested trajectories, indicating local controllability and observability of the linearized dynamics.
In hybrid classical-quantum computing, variational quantum algorithms (VQAs) have emerged as a promising approach in the Noisy Intermediate-Scale Quantum (NISQ) era; however, their performance is often hindered by barren plateaus, where gradients vanish exponentially, rendering optimization ineffective. In this work, we introduce a residual-inspired quantum circuit architecture that incorporates a single messenger qubit, referred to as Q-LINK. By conducting numerical simulations on random quantum states, we observe that Q-LINK significantly enhances optimization behavior by sustaining larger gradient variance and accelerating convergence. Additionally, Q-LINK improves convergence efficiency by 4-6 times and increases gradient variance by up to two orders of magnitude compared with the Vanilla model. To further characterize the impact of the proposed structure, we analyze the expressibility of the circuits before and after introducing Q-LINK and find that the overall expressibility value remains largely unchanged, indicating that the original representational capacity of the circuit is preserved. In addition, we visualize the loss landscapes of different architectures to provide insights into how the proposed design reshapes the cost function landscape. These results demonstrate that introducing only a single messenger qubit can effectively mitigate barren plateau effects while maintaining the ability to explore the Hilbert space of variational quantum circuits.
This paper examines the IDM microscopic car-following model from a dynamical systems perspective, analyzing the effects of delay on congestion formation. Further, a case of mixed-autonomy is considered by controlling one car with Followerstopper in a ring road setting containing IDM vehicles as human drivers. Specifically, the stop-and-go waves phenomenon in idealized traffic from a dynamical systems perspective is examined. We show that Followerstopper-controlled vehicle is effective at eliminating emergent stop-and-go waves in the IDM traffic simulation. We show through simulation that the uniform flow manifold is unstable for the ring road simulation with IDM vehicles, and that replacing a single car with Followerstopper induces stability, allowing the cars to drive safely at a uniform speed. Additionally, the case of known delay is considered in a mixed-autonomy scenario. Our simulation result shows that while considering a known time delay, traffic waves emerge earlier than in the no-delay case. At the same time, a single-vehicle controlled using Followerstopper controller is able to prevent the emergence of traffic waves even in the presence of delay.
Barren plateaus, which means the training gradients become extremely small, pose a major challenge in optimizing parameterized quantum circuits, often making the learning process impractically slow or stall. This work shows why using neural networks to generate quantum circuit parameters helps overcome this difficulty. We introduce a geometric viewpoint that describes how the parameters produced by neural networks evolve during training. Our analysis shows that these parameters follow smooth and efficient paths that avoid the flat regions in the training that cause barren plateaus. This provides a computational explanation for the improved trainability observed in recent neural network-assisted quantum learning methods. Overall, our findings bridge ideas from quantum machine learning and computational optimization, offering new insight into the structure of quantum models and guiding future approaches for designing more trainable quantum circuits or parameter initialization.
Variational quantum algorithms (VQAs) combine the advantages of classical optimization and quantum computation, making them one of the most promising approaches in the Noisy Intermediate-Scale Quantum (NISQ) era. However, when optimized using gradient descent, VQAs often suffer from the vanishing gradient problem, commonly known as the barren plateau. Various methods have been proposed to mitigate this issue. In this work, we propose a hybrid approach that integrates a classical proportional-integral-derivative (PID) controller with a neural network to update the parameters of variational quantum circuits. We refer to this method as NPID, which aims to mitigate the barren plateau. The proposed algorithm is tested on randomly generated quantum input states and random quantum circuits with parametric noise to evaluate its universality, and additional simulations are conducted under different noise rates to examine its robustness. The effectiveness of the proposed method is evaluated based on its convergence speed toward the target cost value. Simulation results show that NPID achieves a convergence efficiency 2-9 times higher than NEQP and QV, with performance fluctuations averaging only 4.45% across different noise levels. These results highlight the potential of integrating classical control theory into quantum optimization, providing a new perspective for improving the trainability and stability of variational quantum algorithms.
We present a load frequency control strategy deploying a decentralized robust global integral terminal sliding mode control (GITSMC) method to maintain stable frequency and tie-line power in multi-area interconnected microgrids with aggregated uncertainties. To achieve this, firstly, we have developed a mathematical model of the multi-area interconnected system incorporating disturbances from solar photovoltaic (PV), wind turbine (WT) generation and load demand, as aggregated uncertainties. Secondly, we have designed a global integral terminal sliding surface with an exponential reaching law for each area to enhance system dynamic performance and suppress chattering within a finite time. Thirdly, the overall stability of the closed-loop system is analyzed using the Lyapunov stability theorem. Finally, extensive simulations are conducted on the IEEE 10-generator New England 39-bus power system, including load disturbances and variable PV and WT generation. The results demonstrate the performance of the proposed GITSMC approach, achieving approximately 94.9% improvement in ITSE and 94.4% improvement in ISE, confirming its superior accuracy and dynamic performance compared to the existing controller.
The increasing adoption of electric vehicles (EVs) necessitates an understanding of their driving behavior to enhance traffic safety and develop smart driving systems. This study compares classical and machine learning models for EV car following behavior. Classical models include the Intelligent Driver Model (IDM), Optimum Velocity Model (OVM), Optimal Velocity Relative Velocity (OVRV), and a simplified CACC model, while the machine learning approach employs a Random Forest Regressor. Using a real world dataset of an EV following an internal combustion engine (ICE) vehicle under varied driving conditions, we calibrated classical model parameters by minimizing the RMSE between predictions and real data. The Random Forest model predicts acceleration using spacing, speed, and gap type as inputs. Results demonstrate the Random Forest's superior accuracy, achieving RMSEs of 0.0046 (medium gap), 0.0016 (long gap), and 0.0025 (extra long gap). Among physics based models, CACC performed best, with an RMSE of 2.67 for long gaps. These findings highlight the machine learning model's performance across all scenarios. Such models are valuable for simulating EV behavior and analyzing mixed autonomy traffic dynamics in EV integrated environments.
When electric vehicles (EVs) are integrated into standalone DC microgrids (DCMGs), stability issues arise due to their constant power load (CPL) behavior, which provides negative incremental impedance (NII). In addition, the microgrids suffer from an inherent low-inertia problem. Therefore, this study presents a composite controller incorporating a global integral terminal sliding mode controller with a backstepping controller. A virtual capacitor is employed to mitigate the low-inertia issue and strengthen the DC-bus response. An improved fractional power-based reaching law decreases chattering and accelerates convergence. Exact feedback linearization converts the nonlinear boost converter model into Brunovsky's canonical form, mitigating NII effects and non-minimum phase issues. The entire system stability is verified using Lyapunov control theory. Simulation outcomes confirm superior performance, with 34.4-53.3% reduction in overshoot, 52.9-74.9% in undershoot, and 12-47.4% in settling time compared to the existing controller.
This paper revisits Followerstopper, a phase-space-based control system that had demonstrated its ability to mitigate emergent traffic jams due to stop-and-go traffic during rush hour in the mixed-autonomy setting. Followerstopper was deployed on an autonomous vehicle. The controller attenuates the emanant traffic waves by regulating its velocity according to the relative distance and velocity of the leader car. While regulating the velocity, the controller also prevents the collision of the ego vehicle with the lead vehicle within the range specified by the controller's design parameter. The controller design is based on a configurable quadratic curve on relative distance-relative velocity phase-space that allows the transition of the regulated velocity from (i) no modification of input, (ii) decelerating to match the leader's velocity (iii) braking to avoid any imminent collision. In this paper, we explore the phase-space properties of Followerstopper and provide a detailed description of a nonlinear control law that regulates the reference input to Followerstopper within the physics-informed boundaries. We also provide a new discussion on the nominal control law that regulates the reference speed to Followerstopper to avoid unrealistic and unsafe acceleration.
We find that using neural networks to generate quantum states can effectively alleviate the barren plateau phenomenon present in random variational quantum circuits.
Current research on trajectory prediction primarily relies on data collected by onboard sensors of an ego vehicle. With the rapid advancement in connected technologies, such as vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication, valuable information from alternate views becomes accessible via wireless networks. The integration of information from alternative views has the potential to overcome the inherent limitations associated with a single viewpoint, such as occlusions and limited field of view. In this work, we introduce V2INet, a novel trajectory prediction framework designed to model multi-view data by extending existing single-view models. Unlike previous approaches where the multi-view data is manually fused or formulated as a separate training stage, our model supports end-to-end training, enhancing both flexibility and performance. Moreover, the predicted multimodal trajectories are calibrated by a post-hoc conformal prediction module to get valid and efficient confidence regions. We evaluated the entire framework using the real-world V2I dataset V2X-Seq. Our results demonstrate superior performance in terms of Final Displacement Error (FDE) and Miss Rate (MR) using a single GPU. The code is publicly available at: https://github.com/xichennn/V2I_trajectory_prediction.
The prediction of surrounding vehicle trajectories is crucial for collision-free path planning. In this study, we focus on a scenario where a connected and autonomous vehicle (CAV) serves as the central agent, utilizing both sensors and communication technologies to perceive its surrounding traffics consisting of autonomous vehicles (AVs), connected vehicles (CVs), and human-driven vehicles (HDVs). Our trajectory prediction task is aimed at all the detected surrounding vehicles. To effectively integrate the multi-source data from both sensor and communication technologies, we propose a deep learning framework called MSMA utilizing a cross-attention module for multi-source data fusion. Vector map data is utilized to provide contextual information. The trajectory dataset is collected in CARLA simulator with synthesized data errors introduced. Numerical experiments demonstrate that in a mixed traffic flow scenario, the integration of data from different sources enhances our understanding of the environment. This notably improves trajectory prediction accuracy, particularly in situations with a high CV market penetration rate. The code is available at: https://github.com/xichennn/MSMA.
With low-cost computing devices, improved sensor technology, and the proliferation of data-driven algorithms, we have more data than we know what to do with. In transportation, we are seeing a surge in spatiotemporal data collection. At the same time, concerns over user privacy have led to research on differential privacy in applied settings. In this paper, we look at some recent developments in differential privacy in the context of spatiotemporal data. Spatiotemporal data contain not only features about users but also the geographical locations of their frequent visits. Hence, the public release of such data carries extreme risks. To address the need for such data in research and inference without exposing private information, significant work has been proposed. This survey paper aims to summarize these efforts and provide a review of differential privacy mechanisms and related software. We also discuss related work in transportation where such mechanisms have been applied. Furthermore, we address the challenges in the deployment and mass adoption of differential privacy in transportation spatiotemporal data for downstream analyses.
Cyber-physical systems such as microgrids consist of interconnected components, localized power systems, and distributed energy resources with clearly defined electrical boundaries. They can function independently but can also work in tandem with the main grid. Power system converters and their control loops play an essential role in stabilizing grids and interfacing a microgrid with the main grid. The optimal selection of microgrid components for installation is expensive. Simulation of microgrids provides a cost-effective solution. However, when studying the electromagnetic transient response, their simulation is slow. Furthermore, software packages facilitating electromagnetic transient response may be prohibitively expensive. This paper presents a faster method for simulating the electromagnetic transient response of microgrid components using SystemC-AMS. We present a use case of a photovoltaic grid-following inverter with a phase-locked loop to track reference active and reactive power. Our results demonstrate that the simulation performed using SystemC-AMS is roughly three times faster than the benchmark simulation conducted using Simulink. Our implementation of a photovoltaic grid-following inverter equipped with a phase-locked loop for monitoring reference active and reactive power reveals that the simulation executed using SystemC-AMS is approximately three times faster than the benchmark simulation carried out using Simulink. Our implementation adopts a model-based design and produces a library of components that can be used to construct increasingly complex grid architectures. Additionally, the C-based nature allows for the integration of external libraries for added real-time capability and optimization functionality. We also present a use case for real-time simulation using a DC microgrid with a constant resistive load.
The use of pre-shared entanglement in entanglement-assisted communication offers a superior alternative to classical communication, especially in the photon-starved regime and highly noisy environments. In this paper, we analyze the performance of several low-complexity receivers that use optical parametric amplifiers. The simulations demonstrate that receivers employing an entanglement-assisted scheme with phase-shift-keying modulation can outperform classical capacities. We present a 2x2 optical hybrid receiver for entanglement-assisted communication and show that it has a roughly 10% lower error probability compared to previously proposed optical parametric amplifier-based receivers for more than 10 modes. However, the capacity of the optical parametric amplifier-based receiver exceeds the Holevo capacity and the capacities of the optical phase conjugate receiver and 2x2 optical hybrid receiver in the case of a single mode. The numerical findings indicate that surpassing the Holevo and Homodyne capacities does not require a large number of signal-idler modes. Furthermore, we find that using unequal priors for BPSK provides roughly three times the information rate advantage over equal priors.
Cyber-physical power systems, such as grids, integrate computational and communication components with physical systems to introduce novel functions and improve resilience and fault tolerance. These systems employ computational components and real-time controllers to meet power demands. Microgrids, comprising interconnected components, energy resources within defined electrical boundaries, computational elements, and controllers, offer a solution for integrating renewable energy sources and ensuring resilience in electricity demand. Simulating these cyber-physical systems (CPS) is vital for grid design, as it facilitates the modeling and control of both continuous physical processes and discrete-time power converters and controllers. This paper presents a model-based design tool for simulating cyber-physical power systems, including microgrids, using SystemC-AMS. The adoption of SystemC-AMS enables physical modeling with both native components from the SystemC-AMS library and user-defined computational elements. We observe that SystemC-AMS can accurately produce the electromagnetic transient responses essential for analyzing grid stability. Additionally, we demonstrate the effectiveness of SystemC-AMS through use cases that simulate grid-following inverters. Comparing the SystemC-AMS implementation to one in Simulink reveals that SystemC-AMS offers a more rapid simulation. A design tool like this could support microgrid designers in making informed decisions about the selection of microgrid components prior to installation and deployment.
The CIRCLES project aims to reduce instabilities in traffic flow, which are naturally occurring phenomena due to human driving behavior. These "phantom jams" or "stop-and-go waves,"are a significant source of wasted energy. Toward this goal, the CIRCLES project designed a control system referred to as the MegaController by the CIRCLES team, that could be deployed in real traffic. Our field experiment leveraged a heterogeneous fleet of 100 longitudinally-controlled vehicles as Lagrangian traffic actuators, each of which ran a controller with the architecture described in this paper. The MegaController is a hierarchical control architecture, which consists of two main layers. The upper layer is called Speed Planner, and is a centralized optimal control algorithm. It assigns speed targets to the vehicles, conveyed through the LTE cellular network. The lower layer is a control layer, running on each vehicle. It performs local actuation by overriding the stock adaptive cruise controller, using the stock on-board sensors. The Speed Planner ingests live data feeds provided by third parties, as well as data from our own control vehicles, and uses both to perform the speed assignment. The architecture of the speed planner allows for modular use of standard control techniques, such as optimal control, model predictive control, kernel methods and others, including Deep RL, model predictive control and explicit controllers. Depending on the vehicle architecture, all onboard sensing data can be accessed by the local controllers, or only some. Control inputs vary across different automakers, with inputs ranging from torque or acceleration requests for some cars, and electronic selection of ACC set points in others. The proposed architecture allows for the combination of all possible settings proposed above. Most configurations were tested throughout the ramp up to the MegaVandertest.