Interactions between T cell receptors (TCRs) and peptide-major histocompatibility complexes (pMHCs) are central to adaptive immunity. Recent advances in structure prediction tools have enabled atomic-level modeling of TCR-pMHC interactions. However, the lack of systematic evaluation forces practitioners to invest substantial resources in selecting appropriate tools. Here, we present a comprehensive benchmark of TCR-pMHC structure prediction with 70 previously unseen complexes and 13 models spanning MSA-based, PLM-based, and docking-based approaches, revealing the superior modeling accuracy and docking quality of MSA-based methods, especially AlphaFold3. To further enhance the utility of AlphaFold3 predictions, we identify the pLDDT score of the TCR CDR3 region as an informative indicator of both structural correctness and functional relevance. Specifically, it enables up to 4.3% Top-1 success gain through reranking and captures mutation-induced affinity changes in 75.3% of cases. Overall, our analysis would facilitate the practical usage of immune structure prediction models and guide the advancement of these models.

This paper focuses on the efficiency analysis and optimization of the Yoke-less Axial-flux Synchronous Motor (YASA), a structured axial flux permanent magnet motor. Firstly, the advantages of the YASA motor, as a new motor topology, are introduced, including high power density, high torque density, efficiency, and a slim design. Secondly, using the YASA motor as an example, a three-dimensional electromagnetic field calculation model is established, and the effects of the slot filling factor, tooth shoe thickness, and tooth shoe width coefficient on motor efficiency are analyzed. By optimizing these design variables, the iron core loss and eddy current loss of the motor are systematically reduced using response surface methodology and genetic algorithms, significantly improving the motor's efficiency. Finally, finite element analysis is used to verify the feasibility of the optimized scheme on a 10-pole, 12-slot motor. The results show that the iron core loss and eddy current loss of the motor are reduced by 36.5% and 37.8%, respectively.

The development of integrated on-board chargers has improved the space utilization and conserved hardware resources in electric vehicles (EVs). However, many integration methods modify the inverter and motor structure, and most methods failed to accommodate variable input voltage levels. Therefore, this paper proposed an integrated on-board charger with Buck and Buck-Boost capability. The charger reused the inverter and the interior permanent magnet synchronous motor (IPMSM) as Buck and Buck-Boost converter. The topology could realize the charging only by connecting the anode of the battery to the output terminal of the A phase of the inverter, thus minimizing changes to the drive system. Besides, due to the different calculation methods of the Buck mode and Buck-Boost mode, a direct switching between the two modes could result in a current shock. To address the impact of this problem, a dual-mode hybrid control algorithm was put forward. The paper analyzed the whole charging process and the principle of the dual-mode hybrid control in detail. Finally, experiments based on the 1.5-kW IPMSM realized the Buck and Buck-Boost charging for supercapacitors.

Ultra-wideband (UWB)-vision fusion localization has achieved extensive applications in the domain of multiagent relative localization. The challenging matching problem between robots and visual detection renders existing methods highly dependent on identity-encoded hardware or delicate tuning algorithms. Overconfident yet erroneous matches may bring about irreversible damage to the localization system. To address this issue, we introduce Mr. Virgil, an end-to-end learning multi-robot visual-range relative localization framework, consisting of a graph neural network for data association between UWB rangings and visual detections, and a differentiable pose graph optimization (PGO) back-end. The graph-based front-end supplies robust matching results, accurate initial position predictions, and credible uncertainty estimates, which are subsequently integrated into the PGO back-end to elevate the accuracy of the final pose estimation. Additionally, a decentralized system is implemented for real-world applications. Experiments spanning varying robot numbers, simulation and real-world, occlusion and non-occlusion conditions showcase the stability and exactitude under various scenes compared to conventional methods. Our code is available at: https://github.com/HiOnes/Mr-Virgil.

Bug reports and patch commits are dramatically increasing for OS kernels, incentivizing a critical need for kernel-level bug reproduction and patch testing. Directed greybox fuzzing (DGF), aiming to stress-test a specific part of code, is a promising approach for bug reproduction and patch testing. However, the existing DGF methods exclusively target user-space applications, presenting intrinsic limitations in handling OS kernels. In particular, these methods cannot pinpoint the appropriate system calls and the needed syscall parameter values to reach the target location,resulting in low efficiency and waste of resources.

The development of integrated on-board charger has improved the space utilization and conserved hardware resources in electric vehicles (EVs). However, many chargers only have buck or boost charging mode. That may cause the EVs to be unable to match different external charging voltage. Therefore, this paper proposes an integrated on-board charger with buck and boost charging function. The charger reconfigures the inverter and motor winding into a Cuk converter. It does not affect the driving function of the motor, meanwhile, that also gives consideration to the charging function. In addition, both the input current and the output current are continuous when boost or buck charging. And two charging modes are proposed. 1) Constant-current charging mode which keeps the battery current constant when charging. 2) Constant-power charging mode which makes the input power constant. The paper analyzes the whole process and control mode of the two charging modes. Finally, simulation realized the fast and healthy charging for supercapacitors.

Summary Traditional applications have been developed for decades. Most of the security research around them have focused on the detection of memory corruption vulnerabilities, such as buffer overflow, double fetch, and integer overflow. On the contrary, logic bugs, a kind of flaws caused by unreasonable application logic, attract much less attention. Files are the most common media for programs to persist their data in the system. As the file owners, programs are responsible for protecting their files from malicious users' tampering by leveraging access control mechanisms. However, if a program configures their access control mechanisms in wrong ways and causes evil users to bypass security checks to access files, there exists a file access control vulnerability. As a branch of logic flaws, file access control vulnerabilities are less popular with researchers. Thus, to mitigate the harm of the file access control vulnerabilities on Windows system, our team conducted first‐step research on them. We first classified file access control bugs into two types and codified some bug patterns. Then we formalized file access control vulnerabilities to propose a scalable detection method and implemented a lightweight analysis system StaticFAC. After evaluating StaticFAC in real‐world Windows software, we discovered 15 0‐day bugs.

Interior permanent magnet synchronous machines (IPMSM) often work under flux weakening state in electrical vehicles for higher speed. But unavoidable faults might occur during that process, after which the power transistors are switched off immediately and the battery packs are charged through the three-phase uncontrolled rectifier composed of six flywheel diodes. That is also renowned as uncontrolled generation (UCG) operation which needs to be well understood to avoid damage to either inverter or battery. In order to achieve more accurate analytical results, this paper proposes an improved steady-state model of the drive system for UCG analysis. On the basis of the model, the parameter dependencies are detailed. The voltage and current characteristics of UCG and the effectiveness of the proposed steady-state model are verified by experimental results.

This paper directly uses the machine windings to discharge the capacitor voltage in the PMSM drive systems when an emergency happens to electric vehicles. The winding-based method needs to control the d, q-axis current in the motor and it requires accurate position information during the whole discharge process. Whereas, the position sensor installed in the motor is likely to fail due to the violent vibration and impact in the emergency. Considering the position sensor faults, this paper proposes a more reliable current control algorithm based on a sliding mode position estimator to bleed the capacitor voltage. The proposed discharge strategy treats the machine as a generator so as to regulate the DC-bus voltage, and takes advantage of the maximum discharge capacity of the drive system and manages to maintain the discharge time within 2.7 s for the studied system. Simulation and experimental results are presented to verify the effectiveness of the proposed algorithm.