Da‐Wei Ding

Towards enhancement for robot-assisted bending performance of accuracy and process using algorithmic improved neural networks

Da‐Wei Ding, Peng-Yu Wang, Si-Qi Liu, Jin Tao et autres

cn (code pays fourni par la source)

SoAD: Safety-Oriented Value Estimation for Enhanced Closed-Loop End-to-End Autonomous Driving

Yinfeng Gao, Deqing Liu, Yupeng Zheng, Qichao Zhang et autres

End-to-end (E2E) autonomous driving systems, which map sensory inputs directly to vehicle planning, have garnered attention for harnessing the potential of data-driven methodologies in motion planning. However, current methods face two limitations that undermine their safety performance in closed-loop driving tasks. First, …

cn (code pays fourni par la source)

PerlAD: Towards Enhanced Closed-Loop End-to-End Autonomous Driving With Pseudo-Simulation-Based Reinforcement Learning

Yinfeng Gao, Qichao Zhang, Deqing Liu, Zhongpu Xia et autres

End-to-end autonomous driving policies based on Imitation Learning (IL) often struggle in closed-loop execution due to the misalignment between inadequate open-loop training objectives and real driving requirements. While Reinforcement Learning (RL) offers a solution by directly optimizing driving goals via reward signals, …

cn (code pays fourni par la source)

PerlAD: Towards Enhanced Closed-loop End-to-end Autonomous Driving with Pseudo-simulation-based Reinforcement Learning

Yinfeng Gao, Qichao Zhang, Deqing Liu, Zhongpu Xia et autres

End-to-end autonomous driving policies based on Imitation Learning (IL) often struggle in closed-loop execution due to the misalignment between inadequate open-loop training objectives and real driving requirements. While Reinforcement Learning (RL) offers a solution by directly optimizing driving goals via reward signals, …

Learning from Mistakes: Post-Training for Driving VLA with Takeover Data

Yinfeng Gao, Deqing Liu, Qichao Zhang, Yupeng Zheng et autres

Current Vision-Language-Action (VLA) paradigms in end-to-end autonomous driving rely on offline training from static datasets, leaving them vulnerable to distribution shift. Recent post-training methods use takeover data to mitigate this by augmenting the dataset with high-quality expert takeover samples, yet they suffer …

PerlAD: Towards Enhanced Closed-loop End-to-end Autonomous Driving with Pseudo-simulation-based Reinforcement Learning

Yinfeng Gao, Qichao Zhang, Deqing Liu, Zhongpu Xia et autres

End-to-end autonomous driving policies based on Imitation Learning (IL) often struggle in closed-loop execution due to the misalignment between inadequate open-loop training objectives and real driving requirements. While Reinforcement Learning (RL) offers a solution by directly optimizing driving goals via reward signals, …

cn (code pays fourni par la source)

Learning from Mistakes: Post-Training for Driving VLA with Takeover Data

Yinfeng Gao, Deqing Liu, Qichao Zhang, Yupeng Zheng et autres

Current Vision-Language-Action (VLA) paradigms in end-to-end autonomous driving rely on offline training from static datasets, leaving them vulnerable to distribution shift. Recent post-training methods use takeover data to mitigate this by augmenting the dataset with high-quality expert takeover samples, yet they suffer …

Fault Detection Based on Time-Delay-Dependent H - / H ∞ Observers for Linear Discrete Systems

Xiaoxue Wu, Na Li, TianQiu Zhang, Da‐Wei Ding

cn (code pays fourni par la source)

Data-Driven Adaptive Output Formation Control of Heterogeneous Multiagent Systems with Unmeasurable Noise

Jie Zhang, Xiaojie Sun, Jian-An Wang, Chao Deng et autres

cn (code pays fourni par la source)

A Self-Refining Framework for Intracranial Primary Tumors Diagnosis

Zishuo Wan, Haibin Wan, Runting Li, Dabiao Zhou et autres

Accurate preoperative MRI diagnosis of intracranial primary tumors is critical for surgical planning and therapeutic decision-making. This study addresses two fundamental limitations of MRI-based diagnosis: the inherent class imbalance and resolution variations. To address these issues, we propose a novel self-refining framework …

cn (code pays fourni par la source)

Optimal Tracking and Regulation Performance of Networked Control Systems With Time Delay and Input Quantization

Da‐Wei Ding

This technical note investigates the optimal tracking and regulation performance of networked control systems (NCSs) with time delay and input quantization. To this end, advanced trade-off performance indices for multi-input multi-output (MIMO) systems are firstly proposed, aiming to identify new design constraints …

cn (code pays fourni par la source)

Tracking Control for Constrained Nonlinear Cyber‐Physical Systems Against Actuator Attacks: A New Low‐Complexity Approach

Zhaoyang Cuan, Yingying Ren, Da‐Wei Ding, Youyi Wang

ABSTRACT In this article, we investigate the tracking control issue for a kind of constrained nonlinear cyber‐physical systems (CPSs) with exogenous perturbations, which are subjected to venomous actuator attacks. By designing a new orientation function and error transformation function, a novel low‐complexity …

cn, sg (code pays fourni par la source)