| 1. |
- Su, Peng, et al.
(author)
-
A Simulation-Aided Approach to Safety Analysis of Learning-Enabled Components in Automated Driving Systems
- 2023
-
In: Proceedings of 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC). - : Institute of Electrical and Electronics Engineers (IEEE).
-
Conference paper (peer-reviewed)abstract
- Artificial Intelligence (AI) techniques through Learning-Enabled Components (LEC) are widely employed in Automated Driving Systems (ADS) to support operation perception and other driving tasks relating to planning and control. Therefore, the risk management plays a critical role in assuring the operational safety of ADS. However, the probabilistic and nondeterministic nature of LEC challenges the safety analysis. Especially, the impacts of their functional faults and incompatible external conditions are often difficult to identify. To address this issue, this article presents a simulation-aided approach as follows: 1) A simulation-aided operational data generation service with the operational parameters extracted from the corresponding system models and specifications; 2) A Fault Injection (FI) serviceaimed at high-dimensional sensor data to evaluate the robustness and residual risks of LEC. 3) A Variational Bayesian (VB) method for encoding the collected operational data and supporting an effective estimation of the likelihood of operational conditions. As a case study, the paper presents the results of one experiment, where the behaviour of an Autonomous Emergency Braking(AEB) system is simulated under various weather conditions based on the CARLA driving simulator. A set of fault types of cameras, including solid occlusion, water drop, salt and pepper, are modelled and injected into the perception module of the AEB system in different weather conditions. The results indicate that our framework enables to identify the critical faults under various operational conditions. To approximate the critical faults in undefined weather, we also propose Variational Autoencoder(VAE) to encode the pixel-level data and estimate the likelihood.
|
|
| 2. |
- Yuan, Ye, et al.
(author)
-
Smart Pavement: An Attention-Based Classification Model for Road Pavement Material
- 2022
-
In: Smart Innovation, Systems and Technologies. - Singapore : Springer Nature Singapore. - 2190-3026 .- 2190-3018. ; 304 SIST, s. 133-140
-
Conference paper (peer-reviewed)abstract
- Intelligent recognition of traffic road damage is essential for realizing smart vehicles and intelligent transportation systems. The classification of road material types before recognition is a challenge for traffic road damage recognition due to differences in features such as concrete and asphalt. In addition, the widely distributed roads make environmental factors a critical factor affecting the classification. In this paper, we propose a deep learning-based road material classification method that introduces an attention mechanism to deal with the influence of different environments on road material recognition. We acquired tens of thousands of road surface images for training and testing and performed practical validation in real roads. The experiments show that our method has high accuracy and recall in road material classification.
|
|
| 3. |
|
|
| 4. |
- Chen, Z., et al.
(author)
-
Plastic Deformation and Residual Stress in High Speed Turning of AD730™ Nickel-based Superalloy with PCBN and WC Tools
- 2018
-
In: Procedia CIRP. - : Elsevier BV. - 2212-8271. ; 71, s. 440-445, s. 440-445
-
Conference paper (peer-reviewed)abstract
- A higher gas turbine efficiency can be achieved by increasing the operating temperature in hot sections. AD730™ is a recently-developed wrought/cast nickel-based superalloy which can maintain excellent mechanical properties above 700. However, machining of AD730™ could be a difficult task like other nickel-based superalloys. Therefore, studies are needed with respect to the machinability of this new alloy. In this paper, high-speed turning was performed on AD730™ using polycrystalline cubic boron nitride (PCBN) tools and coated tungsten carbide (WC) tools at varied cutting speeds. The surface integrity was assessed in two important aspects, i.e., surface and sub-surface plastic deformation and residual stresses. The PCBN tools generally showed better performance compared with the WC tools since it led to reduced machining time without largely compromising the surface integrity achieved. The optimal cutting speed was identified in the range of 200-250 m/min when using the PCBN tools, which gives rise to a good combination of machining efficiency and surface integrity. The further increase of the cutting speed to 300 m/min resulted in severe and deep plastic deformation. Meanwhile, a continuous white layer was formed at the machined surface. When turning with the WC tools, the increased cutting speed from 80 m/min to 100 m/min showed very little effect with respect to the plastic deformation on the machined surface. It was found that tensile residual stresses were developed on all machined surfaces no matter when the PCBN or WC tools were used, and the surface tension was generally increased with increasing cutting speed. The tensile layer might need to be modified by e.g., post-machining surface treatments such as shot peening, if taking good fatigue performance into consideration.
|
|
| 5. |
- Chen, Zhe, et al.
(author)
-
Surface Integrity and Fatigue Performance of Inconel 718 in Wire Electrical Discharge Machining
- 2016
-
In: 3RD CIRP CONFERENCE ON SURFACE INTEGRITY. - : Elsevier BV. ; , s. 307-310
-
Conference paper (peer-reviewed)abstract
- This paper presents a study to characterize the surface integrity in wire electrical discharge machining (EDM) of Inconel 718 and investigate its effect on the fatigue performance of the alloy in a four-point bending fatigue mode at room temperature. The EDM process generates a rough recast surface with multi-types of defects. Surface craters, micro-cracks and micro-voids within the recast layer have been found to be most detrimental from the point of view of fatigue as they could provide many preferential initiation sites for fatigue cracks. As a consequence, the specimens with an EDM cut surface show an approximately 30% decrease in fatigue life compared to those with a polished surface, and multiple crack origins were observed on the fracture surface. The high tensile residual stresses generated on the EDM cut surface, on the other hand, are also believed to be partly responsible for the loss in fatigue life of the alloy machined by EDM.
|
|
| 6. |
- Zhou, Jinming, et al.
(author)
-
Analysis of subsurface microstructure and residual stresses in machined Inconel 718 with PCBN and Al2O3-SiCw tools
- 2014
-
In: Procedia CIRP. - : Elsevier BV. - 2212-8271. ; 13, s. 150-155, s. 150-155
-
Conference paper (peer-reviewed)abstract
- Subsurface microstructural alterations and residual stresses caused by machining significantly affect component lifetime and performance by influencing fatigue, creep, and stress corrosion cracking resistance. Assessing the surface quality of a machined part by characterizing subsurface microstructural alterations and residual stresses is essential for ensuring part performance and lifetime in aero-engines and power generators. This comparative study characterizes and analyzes subsurface microstructural alterations and residual stresses in Inconel 718 subjected to high-speed machining with PCBN and whisker-reinforced ceramic cutting tools. Effects of cutting tool materials and microgeometry on subsurface deformation, microstructural alterations, and residual stresses were investigated. Surface and subsurface regions of machined specimens were investigated using X-ray diffraction, electron channeling contrast imaging, and electron back-scatter diffraction to characterize microstructural alterations and measure deformation intensity and depth. (C) 2014 The Authors. Published by Elsevier B.V.
|
|
| 7. |
- Zhou, J. M., et al.
(author)
-
Surface Characterization of AD730TM Part Produced in High Speed Turning with CBN tool
- 2018
-
In: Procedia CIRP. - : Elsevier BV. - 2212-8271. ; 71, s. 215-220, s. 215-220
-
Conference paper (peer-reviewed)abstract
- AD730TM is a novel superalloy developed for the hot section part in aero engine and gas turbine machinery with enhanced performance. The material is characterized by its excellent high temperature properties for being an alloy possible to manufacture by cast and wrought process compared to other superalloys in the same class such as Inconel718. The material with higher temperature capability means potentially increased energy efficiency as well as less emission in the new engine design. However, there is lack of information on machinability of the material, especially achievable surface quality under high speed machining. Machining process is commonly employed in the manufacturing of hot section part, such as turbine disc, to obtain the final surface quality and tolerance. Surface quality produced by the machining processes is one of the crucial factors to determine the functional performance and correspondent fatigue life time of the parts. The paper will present the results of surface characterization of the part produced by high speed turning with CBN tool. A series machining tests were conducted in the study. Surface morphology on machined samples were investigated with a scanning electron microscope to assess the surface damages and other surface defects induced by the machining processes. 3D surface topology was also analyzed with an Infinite-focus-variant microscope to correlate with machining condition. In addition, effect of tool wear and cutting parameters, such as cutting speed, will also be discussed in the paper.
|
|
| 8. |
- Su, Peng, et al.
(author)
-
Scheduling Resource to Deploy Monitors in Automated Driving Systems
- 2023
-
In: Dependable Computer Systems and Networks. - : Springer Nature. ; , s. 285-294
-
Conference paper (peer-reviewed)abstract
- Deep Neural Networks (DNN) constitute an important technology for operational perception in Automated Driving Systems (ADS). However, the trustworthiness of such DNN is one concern in the system engineering and quality management. Therefore, it is critical to monitor conditions and ensure the safety of the implementations for this advanced technology. One solution is to use Conditional Monitors (CM) to detect possible faults. However, such monitors challenge resource (e.g., data and memory) management of limited memory space in the ADS hardware. This paper proposes a resource scheme for deploying a monitor in ADS by integrating dynamic memory scheduling with Responsibility-Sensitive Safety (RSS). We use the car-following system as a case study to evaluate our scheme. YOLOv5 and KITTI datasets simulate a perception module where various monitors detect faults. We measure the time cost of conventional scheduling pipelines and our method. Compared with the conventional method, our scheme reduces 43.7% of execution time per cycle.
|
|
| 9. |
- Chen, Peng, et al.
(author)
-
Analysis and design of an 1-20 GHz track and hold circuit
- 2021
-
In: 2021 IEEE International Symposium on Circuits and Systems, ISCAS 2021 - Proceedings. - 0271-4310. - 9781728192017 ; 2021-May
-
Conference paper (peer-reviewed)abstract
- This work analyzes the nonlinear effects in the track and hold circuit applied in high-speed ADCs or RF sampling receiver (RX) front-ends. Non-ideal effects inside the main sampling NMOS switch are studied. Parasitic varactor and sampling on-resistance modulation effects are analyzed through frequency domain Volterra series and the EKV MOS transistor model. Polynomial curve fitting is applied showing that the on-resistance modulation dominates. Finally, a novel bootstrap circuit is proposed with a fast settling time and high bootstrap voltage in a 22 nm FD-SOI CMOS technology, with its settling time analyzed using the Elmore delay model.
|
|
| 10. |
- Chen, Y., et al.
(author)
-
Brain injury prediction for vulnerable road users in vehicle accidents using mathematical models
- 2010
-
In: International Federation for Medical and Biological Engineering Proceedings. - Berlin, Heidelberg : Springer Berlin Heidelberg. - 1680-0737. - 9783540790389 ; 31, s. 497-500
-
Conference paper (peer-reviewed)abstract
- The objective was to analyze the brain tissue responses and predict head-brain injuries. Accident reconstructions were carried out by using MBS and FE models based on real-life accident investigation. Thirty cases of VRUs accidents were selected from IVAC and GIDAS databases for simulation study. The brain injury parameters were calculated in terms of coup/countercoup pressure, von Mises and maximum shear stresses at the cerebrum, the callosum, the cerebellum and the brain stem. The correlation of calculated parameters was determined with injury codes observed in accident data. The results indicated that peak coup/countercoup pressures often occur at the cerebrum, while von Mises and maximum shear stresses are both concentrated at the upper end of the brain stem. Physical parameters employed in this study are capable of predicting brain injuries. © 2010 International Federation for Medical and Biological Engineering.
|
|
