Virtual car models for handling and ride bridging off and on-line simulations

Transcription

1 Virtual car models for handling and ride bridging off and on-line simulations Dr.Harald Wilhelm, Simulation Fahrverhalten Audi AG Diego Minen, Technical Director VI-Grade Vehicle EPO Stuttgart VI-CarRealTime as Part of the Audi Chassis Simulation Environment Executive Summary AUDI AG Current Simulation Strategy for Virtual Handling and Ride Dynamics VI-grade Current Tools and Development Strategy for bridging off-line and on-line Simulation 2 1

2 Chassis Control Systems Audi Dynamic Steering Active Differential Semi-active Damper Airspring? Powertrain Control ESP including ABS, EBV, ASR, MSR? interaction must be represented in simulation model 3 Tasks of Simulation Environments PP PF LH1 LF TPPA SOP Vehicle Properties P r o d u k t d e f i n i t i o n P r o d u k t e n t s t. K o m p o n e n t e n e n t w i c k l u n ED KF g SiL/MiL VPT P r o d u k t e n t s t e h u n g S e r i e n e n t w i c k l u n g / S e r i e n v o r b e r e i t u n g HiL Monitoring of Characteristics EF Module Function, Properties Component Design, Function Development of Characteristics SiL/MiL HiL ViL Validation of Characteristics 4 2

3 MATLAB Integration Model (MiL&SiL) Mechanical Vehicle Sensors Damper Control Actuators 5 Mechanical Vehicle Modeling Complex Reference Model: ADAMS/Car Handling Model (Pac2002, Non-Lin Bushings, Flex Bodies ) ADAMS/Car Ride Model (FTire, Hydro&Freq Bushings, Flex Bodies ) Simplified Model: VI-CRT Model for Handling MATLAB Model for Ride (AUDI development) 6 3

4 Mechanical Vehicle Preprocessing Scripted Automation in ADAMS/Car ADAMS/Car Handling Model Scripted Automation in ADAMS/Car ADAMS/Car Ride Model manual VI-CarRealTime-Model MATLAB Vertical Model 7 Examples of Simulation Tasks Task Handling Development Ride Development Powertrain Effects on Handling&Ride Durability Design of Control Algorithms Body-To-Ground Clearance Specifications for Actuators and Sensors Support of Component Parametrization Robustness Check of Vehicle Configurations Lap Time Reference Model Fast Model 8 4

5 Development Loops ext Loop start int Loop start int automatic Loop start int automatic Loop end int Loop end ext Loop end ACAR passive vehicle model & simulation VI-CRT active vehicle model & simulation VI-CRT active vehicle + + = Optimal Design 9 Comparison ADAMS vs. VI-CRT Setups Anti Rollbars Outer Diameter Base Front [mm] 12.8 Rear [mm] 9 Modification Modification

6 Damper Control System Example Mechanical Vehicle Sensors Damper Control Actuators 11 Settings of the Controlled Damper Mechanical Car: ADAMS Mechanical Car: VI-CRT Full A4 Comfort Full A4 Auto Full A4 Dynamic Step Steer (100km/h) CRT A4 Comfort CRT A4 Auto CRT A4 Dynamic Step Steer (100km/h) Roll Velocity [deg/s] Roll Velocity [deg/s] Time [s] Time [s] 12 6

7 Simpert and VI-CarRealTime: future Scripted Automation Car Handling Model Car Ride Model Scripted Automation VI-CRT Model with Powertrain Mass Motion 13 VI-CarRealTime "ADAMS goes RealTime" Full compatibility with ADAMS/Car: Common data model Same driver control file Same data output format Vehicle dynamics conceptual multi-objective optimization Control systems development VI-grade Software, Hardware and Model in-the-loop 20+ years experience with ADAMS development Great understanding of vehicle dynamics Collaboration with major international OEM s 14 7

8 VI-CarRealTime Product Strategy Vehicle Dynamics, Ride, Durability, Fuel Consumption Software-in-the-loop DOE & Optimization Benchmarking Hardware-in-the-loop 15 VI-CarRealTime Components Vehicle Generation Susp Curves Susp Loadcase Files Property files Susp Curves Generation Road data generation VI-Vehicle VI-SuspGen VI-Road ADAMS/Car plugin Model through socket or files Model through socket or files VI-Solver VI-Driver HiL, HuiL 16 Matlab SiL/MiL VI-Animator 8

9 VI-Grade HIL&Driving Simulator Cockpit Virtual Dashboard Vehicle Model Real-time Machine ECU 17 Human-in-the-Loop More and more control systems are included in a car to facilitate drivability, safety and comfort Numerical modelling of these systems has reached unprecedented levels of accuracy Some vehicle control systems require the human-in-the-loop (ABS, ESP, ASP) how does the human perceive, memorize and learn the system control actions? Are those controls action suitable for the human? Can we inlcude a virtual human model in the DS platform to understand all that? 18 9

10 Sensory Conflict Theory Stimuli Receptors Central Nervous System Responses Motion System Volitional and Reflex Motions Active Motion Internal Model Neural Storage of Expected Signals Eyes Motion Stimuli Vestibulum compare Signal Mismatch Update Internal Model treshold CNS mediating MS sympthoms Motion Sickness Syndrome Proprioception Passive Motion adapted from: Benson C.M., Motion sickness, in K. Ernsting (Ed.) Aviation Medicine, Second ed., Butterworths, London, Simulate the simulator A driving simulator has to reproduce the angular rates and forces that the driver would feel if the simulation was real, but inertia & workspace inertia & workspace The vehicle dynamics signals have to be filtered due to limited workspace WASHOUT FILTER needed PERCEPTION ON VEHICLE + DIFFERENCE TO MINIMIZE - WASHOUT FILTER PERCEPTION ON PLATFORM 20 10

11 Integrated Simulation Solution Consistent framework for the development and execution of multiple, multi-rate, real-time simulations for: legacy code models (VI-CRT) MATLAB/SL Flexible Integrated Simulation Solution: for all Concurrent ihawk 32-bit Multiprocessor Systems with RedHawk Linux relies on RedHawk Real-Time features supports large selection of COTS I/O boards 21 Integrated Simulation Solution Simulink SIMulation Workbench FBS* trigger Legacy code RCIM** I/O boards ihawk * FBS = Frequency Based Scheduler ** RCIM = Real-Time Clock and Interrupt Module RedHawk Linux 22 11

12 Flexible Model-in-the-Loop MiL Simulink SIMulation Workbench Legacy code trigger FBS SiL/MiL RCIM HiL I/O boards ihawk HuiL RedHawk Linux simulation becomes a hybrid reality 23 Integrated Simulation Solution Read Inputs Finalise signal and parameter values for this cycle Simulation Cycle Execute models Write outputs Test Scripts Simulink Model Z SIMulation Model Y1 Workbench Legacy code Model Y2 AI DI ARINC 429 AFD Resolver RVDT CANbus FlexRay Model In-Memory Real-Time Data Base (RTDB) AO DO ARINC 429 AFD Resolver RVDT CANbus FlexRay H.M. I. Admin. Interface Record/Playback Program Interface 24 Operator Data Logger 12

13 Development Loops ext Loop start ACAR passive vehicle model & simulation int Loop start int automatic Loop start int automatic Loop end int Loop end VI-CRT active vehicle model & simulation VI-CRT active vehicle + + = Optimal Design ext Loop end 25 VI-CarRealTime as Part of the Audi Chassis Simulation Environment VI-Green 26 13