In this paper, we consider the learning of a Reduced-Order Linear Parameter-Varying Model (ROLPVM) of a nonlinear dynamical system based on data. This is achieved by a two-step procedure. In the first step, we learn a...
详细信息
In this paper, we consider the learning of a Reduced-Order Linear Parameter-Varying Model (ROLPVM) of a nonlinear dynamical system based on data. This is achieved by a two-step procedure. In the first step, we learn a projection to a lower dimensional state-space. In step two, an LPV model is learned on the reduced-order state-space using a novel, efficient parameterization in terms of neural networks. The improved modeling accuracy of the method compared to an existing method is demonstrated by simulation examples.
This paper introduces an information presentation strategy for pedestrians, aiming to enhance traffic efficiency in a mixed pedestrian-automated vehicle environment, such as a public road. While automated driving tech...
This paper introduces an information presentation strategy for pedestrians, aiming to enhance traffic efficiency in a mixed pedestrian-automated vehicle environment, such as a public road. While automated driving technology has made remarkable progress, interactions with pedestrians on regular roads have mostly been studied in virtual environments using virtual reality goggles. According to these studies, potential traffic efficiency and safety issues arise from pedestrians' limited understanding of automated vehicle behavior. To address this, we propose a human-machine interface employing a head-mounted display (HMD) to mitigate traffic efficiency degradation caused by pedestrians. The proposed system draws upon behavioral economics principles to encourage pedestrians to modify their behavior and develop better interactions with automated vehicles. Simulations were conducted to identify an information presentation strategy that strongly supports learning, and its effectiveness was further validated through experiments involving a real vehicle. Notably, the experimental results confirmed that the information presentation strategies proven effective in simulations also facilitated pedestrian learning during real-world interactions.
Over the past decade, the field of assisted and autonomous driving has experienced significant advancements. However, autonomous driving systems are still challenged by the complexities of dynamic urban environments, ...
详细信息
ISBN:
(数字)9798350342291
ISBN:
(纸本)9798350342307
Over the past decade, the field of assisted and autonomous driving has experienced significant advancements. However, autonomous driving systems are still challenged by the complexities of dynamic urban environments, especially when it comes to predicting and responding to the often stochastic behavior of pedestrians. Current approaches largely concentrate on most likely predictions but tend to ignore their inherent probabilistic nature. Our research introduces a novel Transformer-based multimodal probabilistic prediction model that utilizes a Gaussian Mixture Model (GMM). This approach is simpler than its predecessors, yet it maintains competitive performance, capable of inferring prediction uncertainties using GMM parameters. Additionally, we demonstrate how our prediction model can be incorporated into a risk-aware behavior planner, based on the Chance-Constrained Stochastic Shortest Path (CC-SSP) framework. This planner uses probabilistic trajectory predictions as a Markov transition function to modulate the speed of the autonomous vehicle, effectively keeping the probability of collision below a defined threshold. Our implementation is available at https://***/Murdism/Probabilistic_Pedestrian_Trajectory_***.
This paper presents a control strategy based on a new notion of time-varying fixed-time convergent control barrier functions (TFCBFs) for a class of coupled multi-agent systems under signal temporal logic (STL) tasks....
详细信息
In actual conversational scenarios, we can often determine which parts of the previous dialogue are more critical based on the current inquiry. However, the existing contextual modeling methods often encode the query ...
详细信息
ISBN:
(数字)9781665410205
ISBN:
(纸本)9781665410212
In actual conversational scenarios, we can often determine which parts of the previous dialogue are more critical based on the current inquiry. However, the existing contextual modeling methods often encode the query sentence and the dialogue history in a unified manner, which fails to effectively highlight the inference effect of the query sentence. Moreover, these methods typically process the dialogue history only at the information extraction level, neglecting the treatment of the context itself. In this paper, we propose a novel conversational context modeling technique called DialogNTM. Based on the guidance of the query sentence, the technology can effectively eliminate redundant information by reconstructing the representation of the context. Specifically, we have tweaked the memory and input flow of the Neural Turing Machine (NTM) to encode contextual information in memory and guide the read, write, and erase operations of memory through query sentence. This design simulates the human brain's dynamic retrieval and renewal mechanism of previous memories when dealing with current problems. We have conducted extensive experiments on three publicly available datasets to verify the effectiveness of the DialogNTM model. Compared to the benchmark model, DialogNTM showed significant performance improvements ranging from 11% to 73% across multiple automated evaluation metrics (3.52% to 8.68% in absolute terms).
Power transformers are among the most important assets in the power transmission and distribution grid. However, they suffer from degradation and possible faults causing major electrical and financial losses. Partial ...
Power transformers are among the most important assets in the power transmission and distribution grid. However, they suffer from degradation and possible faults causing major electrical and financial losses. Partial discharges (PDs) are used to identify the insulation health status and their degradation level. PDs are incipient, low-magnitude faults caused by localized dielectric breakdown. Those activities emit signals in many forms, including electrical, chemical, acoustic, electromagnetic, and optical, facilitating various detection methods. This paper provides a theoretical basis for the condition evaluation of an oil-filled power transformer and clarifies the relationship between the operating voltage, void location, and electric-field intensity within the void. This was achieved by investigating the propagation characteristics of partial discharge signals in an oil-filled power transformer using a 3D finite element method (FEM) based simulation. Moreover, the characterization of simulated PD sources at different positions is investigated in this paper. The simulation results are curried out to show that air voids near the windings are subject to greatest peak electric field intensity.
This work proposes a novel and provably correct method for three-dimensional optimal motion planning in complex environments. Our approach models the 3D motion planning problem by solving streamlines of the potential ...
详细信息
ISBN:
(数字)9798350377705
ISBN:
(纸本)9798350377712
This work proposes a novel and provably correct method for three-dimensional optimal motion planning in complex environments. Our approach models the 3D motion planning problem by solving streamlines of the potential fluid flow, filling a gap in traditional motion planning techniques by guaranteeing a closed-loop, smooth and natural-looking navigation solution. Special emphasis is given to an inherent challenge of artificial potential field (APF) methods, namely establishing proofs of safety and stability over the entire optimization process. A model-based actor-critic reinforcement learning algorithm is introduced to approximate the optimal solution to the Hamilton-Jacobi-Bellman equation and update the controller parameters in a deterministic manner. Through a series of ROS-Gazebo software-in-the-loop simulations the proposed methodology demonstrates robustness and outperforms widely used methods such as the RRT
∗
, highlighting its contribution to the field of 3D optimal motion planning.
This paper applies the proposed hybrid force and position control method to the physical robot system with interaction tasks to further improve our previous study. In the control scheme, the variable stiffness based o...
详细信息
ISBN:
(数字)9798331517519
ISBN:
(纸本)9798331517526
This paper applies the proposed hybrid force and position control method to the physical robot system with interaction tasks to further improve our previous study. In the control scheme, the variable stiffness based on proportional integral derivative(PID) admittance control is adopted for interaction force tracking and the radial basis function neural network(RBFNN) based fixed-time control is designed to ensure position tracking. We have performed interaction tasks based on a Baxter robot for drawing on the plane and slope plane with different expected interaction forces and position trajectories. The experiment results indicate that the method performs well in terms of interaction force and trajectory tracking.
In this paper, the problem of fault estimation and localization in the connecting dynamic elements of distributed heating and cooling systems are treated. The fault represents the physical parameter change related to ...
详细信息
ISBN:
(纸本)9781450397117
In this paper, the problem of fault estimation and localization in the connecting dynamic elements of distributed heating and cooling systems are treated. The fault represents the physical parameter change related to the heat transfer between the system and the external environment. First, based on the bi-linear dynamic state space model of the system in the presence of a fault, structural observability analysis using Signed Directed Graph (SDG) has been performed to investigate the sensor placement problem. Then, a nonlinear observer with a parameter adaptation algorithm was proposed for fault estimation. The simulation results show that it can successfully detect and estimate the fault. Fault localization along the length of the element has also been attempted, but it has been found that the localization cannot be performed using practically changeable input variables. Frequency domain analysis is presented to discuss this phenomenon.
The goal of this research is to lay the groundwork for understanding the particle swarm optimization technique. The applicability of this technique is proven by solving a variety of Constraint and unconstraint optimiz...
详细信息
暂无评论