Holzmann Adaptive Cooperation between Driver and Assistant System

Improving Road Safety
1. Auflage 2007
ISBN: 978-3-540-74474-0
Verlag: Springer Berlin Heidelberg
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 225 Seiten

One of the next challenges in vehicular technology field is to improve drastically the road safety. Current developments are focusing on both vehicle platform and diverse assistance systems. This book presents a new engineering approach based on lean vehicle architecture ready for the drive-by-wire technology. Based on a cognitive functionality split, execution and command levels are detailed. The execution level centralized over the stability control performs the motion vector coming from the command level. At this level the driver generates a motion vector which is continuously monitored by a virtual co-pilot. The integration of assistance systems in a safety relevant multi-agent system is presented here to provide first an adequate feedback to the driver to let him recover a dangerous situation. Robust strategies are also presented for the intervention phase once the command vehicle has to be optimized to stay within the safety envelope.

Autoren/Hrsg.

Holzmann, Frédéric

Inhaltsverzeichnis

1;Content;7
2;Part I New concept of cooperation;12
2.1;Needs of improved assistant systems;13
2.1.1;Analysis of the cause of accidents on the road;13
2.1.2;Autonomous vehicles as possible solution;15
2.1.3;Ways for improving the driving safety;16
2.1.3.1;Improvement of the infrastructures;17
2.1.3.2;Improvement of the driver capacities;17
2.1.3.3;Improvement of the vehicles;17
2.1.4;Introduction of assistant systems and inherent problems;18
2.1.4.1;Current integration of assistant systems;18
2.1.4.2;New issues coming from assistant systems;19
2.1.4.3;Behavioral changes with the human supervision;19
2.1.4.4;Risks of complacency;19
2.1.5;Problem statement and improvements with SPARC;20
2.2;Adaptive cooperation between driver and assistant system;21
2.2.1;Vehicle architecture matching the driver cognition flow;21
2.2.2;Horizontal layering integrated into the vehicle;24
2.2.3;Overall presentation of the new concept;25
2.2.4;Presentation of the concept of adaptive cooperation;26
3;Part II Executive level as vehicle platform;30
3.1;Requirements for the executive level;31
3.1.1;Tasks of the executive level;31
3.1.2;Integration of the predictive vehicle dynamics model;32
3.1.2.1;Selection of methodology for the prediction of the vehicle dynamics;32
3.1.2.2;Integration of the predictive vehicle dynamics model;32
3.2;Road--tire friction coefficient estimation;35
3.2.1;Analysis of the current methodologies;35
3.2.2;Predictive camera-based measurement;36
3.2.2.1;Extraction of the ranges analysis;38
3.2.2.2;Statistical approach;38
3.2.2.3;Macroscopic approach;41
3.2.3;Local microphone-based measurement;48
3.2.3.1;Measuring the loud-speaker effect of the tire;48
3.2.3.2;Frequencies extraction from the collected data;49
3.2.3.3;Construction of models;51
3.2.3.4;Matching of the measures;53
3.2.4;Local control of the predictive measures;55
3.2.5;Pros and cons of the estimation methodology;56
3.3;Actuators and drive train architecture;58
3.3.1;Migration strategy to a full safe drive-by-wire platform;58
3.3.2;Drive train architecture;61
3.3.3;Electrical integration with mechanical back-up;62
3.3.4;Electrical replication;64
3.4;Vehicle dynamics model;66
3.4.1;Modeling of the actuators;66
3.4.1.1;Modeling the dynamics of a unit with a second-order transfer function;66
3.4.1.2;Non-iterative identification of the dynamics of units with continuous state;67
3.4.1.3;Identification of the dynamics of the retarder;69
3.4.1.4;Iterative identification of the gear and clutch dynamical model;69
3.4.1.5;Non-iterative identification of the differential model;74
3.4.2;Limitation due to electrical power;75
3.4.2.1;Model of maximal available energy;76
3.4.2.2;Optimizing the energy capacity;76
3.4.2.3;Modifying the command to adapt it to the energy level;77
3.4.2.4;Pre-compensation of the physical limitations;78
3.4.3;Dynamics model;79
3.4.3.1;Computation of the propulsive forces;80
3.4.3.2;Computation of the vehicle dynamics;81
3.4.4;Use of the dynamics model;82
3.5;Performing the vehicle command;84
3.5.1;Command range;84
3.5.2;Inverse computation of the actuators' command;86
3.5.3;Possible extension to a predictive command execution by use of transfer functions;87
3.5.4;Reactive optimization of the command;88
3.5.4.1;Longitudinal correction;88
3.5.4.2;Yaw rate correction with electronic stability control;91
4;Part III Virtual driver for the cooperation;95
4.1;Extended middleware for fault-tolerant architecture;96
4.1.1;Concept of software redundancy with a multi-agent system;96
4.1.2;System management layer;98
4.1.2.1;Agent-based runtime environment;98
4.1.2.2;Use of a blackboard to provide information;100
4.1.2.3;Redundant management of the agents;102
4.1.3;Integration of fail-tolerant agents;108
4.1.3.1;Structure of an agent;108
4.1.3.2;Redundant computation;109
4.2;Agents derived from the robotic field;112
4.2.1;Potential field approach;112
4.2.1.1;Rejection forces;112
4.2.1.2;Lane keeping;114
4.2.1.3;Temporary destination setting;114
4.2.1.4;Resulting acceleration;114
4.2.1.5;Resulting problem;115
4.2.2;Modified dynamic window;116
4.2.2.1;Road monitoring;117
4.2.2.2;Object monitoring;118
4.2.2.3;Fusion of the two sub-modules;120
4.3;Tactic agent for speedway/highway;122
4.3.1;Fusion of reactive and anticipatory action;122
4.3.1.1;Environment categorization;123
4.3.1.2;Choice of the longitudinal and lateral controllers;125
4.3.2;Longitudinal controllers;126
4.3.2.1;Safety acceleration for the front direction;126
4.3.2.2;Distance control for the front direction;127
4.3.3;Lateral controllers;128
4.3.3.1;Safety range for the lane keeping;128
4.3.3.2;Extreme lane keeping assistant for other lanes;134
4.3.3.3;Safety distance for the lane changing;134
4.3.4;Anticipatory action to prevent inappropriate speed;64
4.3.4.1;Computation of the maximal safe speed;137
4.3.4.2;Extension to multiple paths;140
5;Part IV Adaptive cooperation;145
5.1;Methodology of a fault-tolerant adaptive cooperation;146
5.1.1;Drawbacks of current emergency brake;146
5.1.2;Concept of the adaptive cooperation;147
5.1.3;Functionalities degradation by use of recovery blocks;149
5.2;Understanding the driver maneuver;152
5.2.1;A priori choices by looking at the history;152
5.2.2;Weighting the choices with the command dynamics;154
5.2.3;Auto-adaptive detection;156
5.2.3.1;Analysis of the probabilistic graph of the maneuver detection;156
5.2.3.2;Updating the history;157
5.3;Determination of the driver drowsiness;158
5.3.1;Driver and his/her condition;158
5.3.2;Direct non-obtrusive measurement of the drowsiness;159
5.3.2.1;Methodology;159
5.3.2.2;Problem of reliability;160
5.3.3;Combination of multiple indirect measures;161
5.3.3.1;Simulation of test drives;161
5.3.3.2;From measures to indicators;163
5.3.3.3;Setting up of drowsiness references;165
5.3.3.4;Combination of the drowsiness indicators;166
5.3.3.5;Following the drowsiness evolution;167
5.4;Cooperation at the command level;170
5.4.1;Binary intervention;170
5.4.1.1;Concept of intervention;170
5.4.1.2;Meshing algorithm;171
5.4.1.3;Computation of the path transition;174
5.4.1.4;Transition control;176
5.4.1.5;Critical analysis;177
5.4.2;Fuzzy control;178
5.4.2.1;System confidence value;178
5.4.2.2;Adaptive weighting;179
5.4.2.3;Critical analysis;180
5.4.3;Adaptive cooperation;180
5.4.3.1;Concept of accepted dangerousness;181
5.4.3.2;Extension by use of the accepteddangerousness;181
5.4.3.3;Goal-based substitution process;184
5.4.3.4;Event-triggered intervention process;184
5.4.3.5;Fusion of both processes;186
5.4.4;Results and analysis;187
5.5;Feedback management for the driver and the virtual driver;188
5.5.1;Analogy to the delphi method;188
5.5.2;Detection of partial and full conflict situations;189
5.5.3;Feedback to the driver;191
5.5.3.1;Generation of a feedback for the driver;191
5.5.3.2;Different used channels;193
5.5.3.3;Feedback dispatching;194
5.5.4;Feedback to the virtual driver;197
5.5.4.1;Check of conflict due to lane detection;197
5.5.4.2;Check of conflict due to road-user detection;199
5.5.5;Critics on the new feedback extensions;206
6;Part V Discussion on the proposed concept;207
6.1;Concept summary and overview of the functionalities;208
6.1.1;Needs to help the driver in his/her task;208
6.1.2;New vehicle architecture concept;208
6.1.3;Creation of an extended executive level;209
6.1.4;Integration of a virtual driver;211
6.1.5;Concept of adaptive cooperation;212
6.1.6;Results and next steps;214
6.2;General conclusion;215
7;References;217
8;Index;225