| New concept of cooperation | |
| Needs of improved assistant systems | p. 3 |
| Analysis of the cause of accidents on the road | p. 3 |
| Autonomous vehicles as possible solution | p. 5 |
| Ways for improving the driving safety | p. 6 |
| Improvement of the infrastructures | p. 7 |
| Improvement of the driver capacities | p. 7 |
| Improvement of the vehicles | p. 7 |
| Introduction of assistant systems and inherent problems | p. 8 |
| Current integration of assistant systems | p. 8 |
| New issues coming from assistant systems | p. 9 |
| Behavioral changes with the human supervision | p. 9 |
| Risks of complacency | p. 9 |
| Problem statement and improvements with SPARC | p. 10 |
| Adaptive cooperation between driver and assistant system | p. 11 |
| Vehicle architecture matching the driver cognition flow | p. 11 |
| Horizontal layering integrated into the vehicle | p. 14 |
| Overall presentation of the new concept | p. 15 |
| Presentation of the concept of adaptive cooperation | p. 16 |
| Executive level as vehicle platform | |
| Requirements for the executive level | p. 23 |
| Tasks of the executive level | p. 23 |
| Integration of the predictive vehicle dynamics model | p. 24 |
| Selection of methodology for the prediction of the vehicle dynamics | p. 24 |
| Integration of the predictive vehicle dynamics model | p. 25 |
| Road-tire [mu] friction coefficient estimation | p. 27 |
| Analysis of the current methodologies | p. 27 |
| Predictive camera-based measurement | p. 28 |
| Extraction of the ranges analysis | p. 30 |
| Statistical approach | p. 30 |
| Macroscopic approach | p. 33 |
| Local microphone-based measurement | p. 40 |
| Measuring the loud-speaker effect of the tire | p. 40 |
| Frequencies extraction from the collected data | p. 41 |
| Construction of models | p. 43 |
| Matching of the measures | p. 45 |
| Local control of the predictive measures | p. 47 |
| Pros and cons of the estimation methodology | p. 48 |
| Actuators and drive train architecture | p. 51 |
| Migration strategy to a full safe drive-by-wire platform | p. 51 |
| Drive train architecture | p. 54 |
| Electrical integration with mechanical back-up | p. 55 |
| Electrical replication | p. 57 |
| Vehicle dynamics model | p. 59 |
| Modeling of the actuators | p. 59 |
| Modeling the dynamics of a unit with a second-order transfer function | p. 59 |
| Non-iterative identification of the dynamics of units with continuous state | p. 60 |
| Identification of the dynamics of the retarder | p. 62 |
| Iterative identification of the gear and clutch dynamical model | p. 62 |
| Non-iterative identification of the differential model | p. 67 |
| Limitation due to electrical power | p. 68 |
| Model of maximal available energy | p. 69 |
| Optimizing the energy capacity | p. 69 |
| Modifying the command to adapt it to the energy level | p. 70 |
| Pre-compensation of the physical limitations | p. 71 |
| Dynamics model | p. 72 |
| Computation of the propulsive forces | p. 73 |
| Computation of the vehicle dynamics | p. 74 |
| Use of the dynamics model | p. 75 |
| Performing the vehicle command | p. 77 |
| Command range | p. 77 |
| Inverse computation of the actuators' command | p. 79 |
| Possible extension to a predictive command execution by use of transfer functions | p. 80 |
| Reactive optimization of the command | p. 81 |
| Longitudinal correction | p. 81 |
| Yaw rate correction with electronic stability control | p. 84 |
| Virtual driver for the cooperation | |
| Extended middleware for fault-tolerant architecture | p. 91 |
| Concept of software redundancy with a multi-agent system | p. 91 |
| System management layer | p. 93 |
| Agent-based runtime environment | p. 93 |
| Use of a blackboard to provide information | p. 95 |
| Redundant management of the agents | p. 97 |
| Integration of fail-tolerant agents | p. 103 |
| Structure of an agent | p. 103 |
| Redundant computation | p. 104 |
| Agents derived from the robotic field | p. 107 |
| Potential field approach | p. 107 |
| Rejection forces | p. 107 |
| Lane keeping | p. 109 |
| Temporary destination setting | p. 109 |
| Resulting acceleration | p. 109 |
| Resulting problem | p. 110 |
| Modified dynamic window | p. 111 |
| Road monitoring | p. 112 |
| Object monitoring | p. 113 |
| Fusion of the two sub-modules | p. 115 |
| Tactic agent for speedway/highway | p. 117 |
| Fusion of reactive and anticipatory action | p. 117 |
| Environment categorization | p. 118 |
| Choice of the longitudinal and lateral controllers | p. 120 |
| Longitudinal controllers | p. 121 |
| Safety acceleration for the front direction | p. 121 |
| Distance control for the front direction | p. 122 |
| Lateral controllers | p. 123 |
| Safety range for the lane keeping | p. 123 |
| Extreme lane keeping assistant for other lanes | p. 129 |
| Safety distance for the lane changing | p. 129 |
| Anticipatory action to prevent inappropriate speed | p. 131 |
| Computation of the maximal safe speed | p. 132 |
| Extension to multiple paths | p. 135 |
| Adaptive cooperation | |
| Methodology of a fault-tolerant adaptive cooperation | p. 143 |
| Drawbacks of current emergency brake | p. 143 |
| Concept of the adaptive cooperation | p. 144 |
| Functionalities degradation by use of recovery blocks | p. 146 |
| Understanding the driver maneuver | p. 149 |
| A priori choices by looking at the history | p. 149 |
| Weighting the choices with the command dynamics | p. 151 |
| Auto-adaptive detection | p. 153 |
| Analysis of the probabilistic graph of the maneuver detection | p. 153 |
| Updating the history | p. 154 |
| Determination of the driver drowsiness | p. 155 |
| Driver and his/her condition | p. 155 |
| Direct non-obtrusive measurement of the drowsiness | p. 156 |
| Methodology | p. 156 |
| Problem of reliability | p. 157 |
| Combination of multiple indirect measures | p. 158 |
| Simulation of test drives | p. 158 |
| From measures to indicators | p. 160 |
| Setting up of drowsiness references | p. 162 |
| Combination of the drowsiness indicators | p. 163 |
| Following the drowsiness evolution | p. 164 |
| Cooperation at the command level | p. 167 |
| Binary intervention | p. 167 |
| Concept of intervention | p. 167 |
| Meshing algorithm | p. 168 |
| Computation of the path transition | p. 171 |
| Transition control | p. 173 |
| Critical analysis | p. 174 |
| Fuzzy control | p. 175 |
| System confidence value | p. 175 |
| Adaptive weighting | p. 176 |
| Critical analysis | p. 177 |
| Adaptive cooperation | p. 177 |
| Concept of accepted dangerousness | p. 178 |
| Extension by use of the accepted dangerousness | p. 178 |
| Goal-based substitution process | p. 181 |
| Event-triggered intervention process | p. 181 |
| Fusion of both processes | p. 183 |
| Results and analysis | p. 184 |
| Feedback management for the driver and the virtual driver | p. 185 |
| Analogy to the delphi method | p. 185 |
| Detection of partial and full conflict situations | p. 186 |
| Feedback to the driver | p. 188 |
| Generation of a feedback for the driver | p. 188 |
| Different used channels | p. 190 |
| Feedback dispatching | p. 191 |
| Feedback to the virtual driver | p. 194 |
| Check of conflict due to lane detection | p. 194 |
| Check of conflict due to road-user detection | p. 196 |
| Critics on the new feedback extensions | p. 203 |
| Discussion on the proposed concept | |
| Concept summary and overview of the functionalities | p. 207 |
| Needs to help the driver in his/her task | p. 207 |
| New vehicle architecture concept | p. 207 |
| Creation of an extended executive level | p. 208 |
| Integration of a virtual driver | p. 210 |
| Concept of adaptive cooperation | p. 211 |
| Results and next steps | p. 213 |
| General conclusion | p. 215 |
| References | p. 217 |
| Index | p. 225 |
| Table of Contents provided by Ingram. All Rights Reserved. |
