Pyramid Bayesian Method for Model Uncertainty Evaluation of Semantic Segmentation in Autonomous Driving

With the fast-developing deep learning models in the field of autonomous driving, the research on the uncertainty estimation of deep learning models has also prevailed. Herein, a pyramid Bayesian deep learning method is proposed for the model uncertainty evaluation of semantic segmentation. Semantic segmentation is one of the most important perception problems in understanding visual scene, which is critical for autonomous driving. This study to optimize Bayesian SegNet for uncertainty evaluation. This paper first simplifies the network structure of Bayesian SegNet by reducing the number of MC-Dropout layer and then introduces the pyramid pooling module to improve the performance of Bayesian SegNet. mIoU and mPAvPU are used as evaluation matrics to test the proposed method on the public Cityscapes dataset. The experimental results show that the proposed method improves the sampling effect of the Bayesian SegNet, shortens the sampling time, and improves the network performance.

Hierarchical Sizing and Power Distribution Strategy for Hybrid Energy Storage System

This paper proposes a hierarchical sizing method and a power distribution strategy of a hybrid energy storage system for plug-in hybrid electric vehicles (PHEVs), aiming to reduce both the energy consumption and battery degradation cost. As the optimal size matching is significant to multi-energy systems like PHEV with both battery and supercapacitor (SC), this hybrid system is adopted herein. First, the hierarchical optimization is conducted, when the optimal power of the internal combustion engine is calculated based on dynamic programming, and a wavelet transformer is introduced to distribute the power between the battery and the SC. Then, the fuel economy and battery degradation are evaluated to return feedback value to each sizing point within the hybrid energy storage system sizing space, obtaining the optimal sizes for the battery and the SC by comparing all the values in the whole sizing space. Finally, an all-hardware test platform is established with a fully active power conversion topology, on which the real-time control capability of the wavelet transformer method and the size matching between the battery and the SC are verified in both short and long time spans.

Review of DC-DC Converter Topologies Based on Impedance Network with Wide Input Voltage Range and High Gain for Fuel Cell Vehicles

The development of fuel cell vehicles (FCVs) has a major impact on improving air quality and reducing other fossil-fuel-related problems. DC-DC boost converters with wide input voltage ranges and high gains are essential to fuel cells and DC buses in the powertrains of FCVs, helping to improve the low voltage of fuel cells and “soft” output characteristics. To build DC-DC converters with the desired performance, their topologies have been widely investigated and optimized. Aiming to obtain the optimal design of wide input range and high-gain DC-DC boost converter topologies for FCVs, a review of the research status of DC-DC boost converters based on an impedance network is presented. Additionally, an evaluation system for DC-DC topologies for FCVs is constructed, providing a reference for designing wide input range and high-gain boost converters. The evaluation system uses eight indexes to comprehensively evaluate the performance of DC-DC boost converters for FCVs. On this basis, issues about DC-DC converters for FCVs are discussed, and future research directions are proposed. The main future research directions of DC-DC converter for FCVs include utilizing a DC-DC converter to realize online monitoring of the water content in FCs and designing buck-boost DC-DC converters suitable for high-power commercial FCVs.

Simulator Coupled with Distributed Co-Simulation Protocol for Automated Driving Tests

To meet the challenges in software testing for automated vehicles, such as increasing system complexity and an infinite number of operating scenarios, new simulation methods must be developed. Closed-loop simulations for automated driving (AD) require highly complex simulation models for multiple controlled vehicles with their perception systems as well as their surrounding context. For the realization of such models, different simulation domains must be coupled with co-simulation. However, widely supported model integration standards such as functional mock-up interface (FMI) lack native support for distributed platforms, which is a key feature for AD due to the computational intensity and platform exclusivity of certain models. The newer FMI companion standard distributed co-simulation protocol (DCP) introduces platform coupling but must still be used in conjunction with AD co-simulations. As part of an assessment framework for AD, this paper presents a DCP compliant implementation of an interoperable interface between a 3D environment and vehicle simulator and a co-simulation platform. A universal Python wrapper is implemented and connected to the simulator to allow its control as a DCP slave. A C-code-based interface enables the co-simulation platform to act as a DCP master and to realize cross-platform data exchange and time synchronization of the environment simulation with other integrated models. A model-in-the-loop use case is performed with the traffic simulator CARLA running on a Linux machine connected to the co-simulation master xMOD on a Windows computer via DCP. Several virtual vehicles are successfully controlled by cooperative adaptive cruise controllers executed outside of CARLA. The standard compliance of the implementation is verified by exemplary connection to prototypic DCP solutions from 3rd party vendors. This exemplary application demonstrates the benefits of DCP compliant tool coupling for AD simulation with increased tool interoperability, reuse potential, and performance.