A High-accuracy Framework for Vehicle Dynamic Modeling in Autonomous Driving

2021 ◽  
Author(s):  
Shu Jiang ◽  
Yu Wang ◽  
Weiman Lin ◽  
Yu Cao ◽  
Longtao Lin ◽  
...  
Keyword(s):  
Dynamic Modeling ◽  
Autonomous Driving ◽  
High Accuracy ◽  
Vehicle Dynamic

A novel approach for experimental identification of vehicle dynamic parameters

2020 ◽  
Vol 234 (10-11) ◽  
pp. 2634-2648
Author(s):  
Di Yao ◽  
Philipp Ulbricht ◽  
Stefan Tonutti ◽  
Kay Büttner ◽  
Prokop Günther
Keyword(s):  
Parameter Identification ◽  
Dynamic Modeling ◽  
Trajectory Optimization ◽  
Dynamic Parameters ◽  
Vehicle Dynamic ◽  
Test Rig ◽  
Vehicle Simulation ◽  
Virtual Test ◽  
Identification Approach ◽  
Measuring Machine

Pervasive applications of the vehicle simulation technology are a powerful motivation for the development of modern automobile industry. As basic parameters of road vehicle, vehicle dynamic parameters can significantly influence the ride comfort and dynamics of vehicle, and therefore have to be calculated accurately to obtain reliable vehicle simulation results. Aiming to develop a general solution, which is applicable to diverse test rigs with different mechanisms, a novel model-based parameter identification approach using optimized excitation trajectory is proposed in this paper to identify the vehicle dynamic parameters precisely and efficiently. The proposed approach is first verified against a virtual test rig using a universal mechanism. The simulation verification consists of four sections: (a) kinematic analysis, including the analysis of forward/inverse kinematic and singularity architecture; (b) dynamic modeling, in which three kinds of dynamic modeling method are used to derive the dynamic models for parameter identification; (c) trajectory optimization, which aims to search for the optimal trajectory to minimize the sensitivity of parameter identification to measurement noise; and (d) multibody simulation, by which vehicle dynamic parameters are identified based on the virtual test rig in the simulation environment. In addition to the simulation verification, the proposed parameter identification approach is applied to the real test rig (vehicle inertia measuring machine) in laboratory subsequently. Despite the mechanism difference between the virtual test rig and vehicle inertia measuring machine, this approach has shown an excellent portability. The experimental results indicate that the proposed parameter identification approach can effectively identify the vehicle dynamic parameters without a high requirement of movement accuracy.

A Snowplowing Resistance Model for Use in Vehicle Dynamic Modeling#

2005 ◽  
Vol 33 (3-4) ◽  
pp. 359-372 ◽  
Author(s):  
Bahram Ravani ◽  
Magomed Gabibulayev ◽  
T. A. Lasky
Keyword(s):  
Dynamic Modeling ◽  
Vehicle Dynamic ◽  
Resistance Model

Dynamic Modeling and Simulation for Battery Electric Vehicles under Inverter Fault Conditions

2011 ◽  
Vol 110-116 ◽  
pp. 3007-3015
Author(s):  
Gwangmin Park ◽  
Byeongjeom Son ◽  
Daehyun Kum ◽  
Seonghun Lee ◽  
Sangshin Kwak
Keyword(s):  
Electric Vehicle ◽  
Dynamic Modeling ◽  
Electric Vehicles ◽  
Dynamic Performance ◽  
Vehicle Speed ◽  
Calibration Data ◽  
Vehicle Dynamic ◽  
Drive Mode ◽  
Modeling Simulation ◽  
Real Vehicle

This paper presents a dynamic modeling, simulation, and analysis of a Battery Electric Vehicle (BEV) according to vehicle dynamic characteristics. Mathematical model variants for the components of BEVs can be modeled and investigated using the Matlab/Simulink software. In order to compare the dynamic performance of BEVs under inverter fault and normal conditions, the CarSim co-simulation platform is configured with real vehicle calibration data. Using this approach, it was possible to quickly check for dynamic performance issues of an electric vehicle without incurring the time delay and cost. The simulation results such as motor output, vehicle speed/acceleration, and propulsion forces are discussed and compared for each drive mode.

scholarly journals Robust Computing for Machine Learning-Based Systems

2020 ◽  
pp. 479-503
Author(s):  
Muhammad Abdullah Hanif ◽  
Faiq Khalid ◽  
Rachmad Vidya Wicaksana Putra ◽  
Mohammad Taghi Teimoori ◽  
Florian Kriebel ◽  
...  
Keyword(s):  
Machine Learning ◽  
Rapid Development ◽  
Autonomous Driving ◽  
High Accuracy ◽  
Security Threats ◽  
Security Vulnerabilities ◽  
Smart Healthcare ◽  
Super Computer ◽  
Temporal Redundancy ◽  
The Impact

AbstractThe drive for automation and constant monitoring has led to rapid development in the field of Machine Learning (ML). The high accuracy offered by the state-of-the-art ML algorithms like Deep Neural Networks (DNNs) has paved the way for these algorithms to being used even in the emerging safety-critical applications, e.g., autonomous driving and smart healthcare. However, these applications require assurance about the functionality of the underlying systems/algorithms. Therefore, the robustness of these ML algorithms to different reliability and security threats has to be thoroughly studied and mechanisms/methodologies have to be designed which result in increased inherent resilience of these ML algorithms. Since traditional reliability measures like spatial and temporal redundancy are costly, they may not be feasible for DNN-based ML systems which are already super computer and memory intensive. Hence, new robustness methods for ML systems are required. Towards this, in this chapter, we present our analyses illustrating the impact of different reliability and security vulnerabilities on the accuracy of DNNs. We also discuss techniques that can be employed to design ML algorithms such that they are inherently resilient to reliability and security threats. Towards the end, the chapter provides open research challenges and further research opportunities.

scholarly journals DSNet: an efficient CNN for road scene segmentation

2020 ◽  
Vol 9 ◽  
Author(s):  
Ping-Rong Chen ◽  
Hsueh-Ming Hang ◽  
Sheng-Wei Chan ◽  
Jing-Jhih Lin
Keyword(s):  
Real Time ◽  
Network Architecture ◽  
Autonomous Driving ◽  
High Accuracy ◽  
Neural Net ◽  
Scene Segmentation ◽  
Convolutional Network ◽  
Advantages And Disadvantages ◽  
Segmentation Accuracy ◽  
Very High

Road scene understanding is a critical component in an autonomous driving system. Although the deep learning-based road scene segmentation can achieve very high accuracy, its complexity is also very high for developing real-time applications. It is challenging to design a neural net with high accuracy and low computational complexity. To address this issue, we investigate the advantages and disadvantages of several popular convolutional neural network (CNN) architectures in terms of speed, storage, and segmentation accuracy. We start from the fully convolutional network with VGG, and then we study ResNet and DenseNet. Through detailed experiments, we pick up the favorable components from the existing architectures and at the end, we construct a light-weight network architecture based on the DenseNet. Our proposed network, called DSNet, demonstrates a real-time testing (inferencing) ability (on the popular GPU platform) and it maintains an accuracy comparable with most previous systems. We test our system on several datasets including the challenging Cityscapes dataset (resolution of 1024 × 512) with an Mean Intersection over Union (mIoU) of about 69.1% and runtime of 0.0147 s/image on a single GTX 1080Ti. We also design a more accurate model but at the price of a slower speed, which has an mIoU of about 72.6% on the CamVid dataset.

Real-Time Simulation Study of Dynamics for Electric Drive Tracked Vehicle Based on Vortex

2012 ◽  
Vol 263-266 ◽  
pp. 595-599
Author(s):  
Bing Li ◽  
Jianhua Zheng ◽  
Yang Hui Zhou ◽  
Li Xi Luo
Keyword(s):  
Real Time ◽  
Electric Drive ◽  
Vehicle Dynamics ◽  
High Accuracy ◽  
Vehicle Dynamic ◽  
Dynamics Model ◽  
Tracked Vehicle ◽  
Real Time Simulation ◽  
Time Simulation ◽  
Vehicle Dynamic Simulation

Aiming at the problem of real-time simulation of vehicle dynamics. Dynamic model of tracked vehicle was built in Vortex. Base class of vehicle was secondary developed to apply torque to the sprockets directly. Finally, dynamics model of electric drive tracked vehicle was established. Under different conditions，the dynamics real-time simulation was carried out. The results showed that the vehicle dynamic simulation in Vortex ensures high accuracy and also has a good real-time.

Full Vehicle Dynamic Modeling for Engine Shake with Hydraulic Engine Mount

2017 ◽  
Author(s):  
Rong Guo ◽  
Jun Gao ◽  
Xiao-kang Wei ◽  
Zhao-ming Wu ◽  
Shao-kang Zhang
Keyword(s):  
Dynamic Modeling ◽  
Vehicle Dynamic ◽  
Full Vehicle ◽  
Engine Mount

Coupling the “Sandwich” Track Model to a Commercial Dynamic Vehicle Model

2014 ◽  
Author(s):  
Shelby Stigers ◽  
Yin Gao ◽  
Hai Huang
Keyword(s):  
Case Studies ◽  
Dynamic Modeling ◽  
Contact Force ◽  
Dynamic Models ◽  
Commercial Vehicle ◽  
Coupling Effect ◽  
Dynamic Interaction ◽  
Vehicle Dynamic ◽  
Vehicle Model ◽  
Simulated Case

Most commercial vehicle dynamic modeling programs do not have, or have a very limited, track component. Today, most track dynamic models do not have a detailed vehicle model either. Assumptions have to be made to simulate the vehicle and track performances when using those programs. For instance, a predefined wheel-rail contact force is normally assumed as the input for both vehicle and track models. This research aims to integrate a previously developed “Sandwich” dynamic track model to a detailed vehicle model in a commercially available program so that a vehicle-track dynamic interaction can be realistically simulated. Case studies will be presented to demonstrate the coupling effect and the results will help us better understand the vehicle and track behavior under different riding conditions.

Sign in / Sign up

Export Citation Format

Share Document