Page 12 - Dynamic Vision for Perception and Control of Motion

Contents xiii

8.2.1 Starting from Zero Curvature for Near Range ........................... 229

8.2.2 Road Curvature from Look-ahead Regions Further Away ........ 230
8.2.3 Simple Numerical Example of Initialization.............................. 231
8.3 Selection of Tuning Parameters for Recursive Estimation.................... 233
8.3.1 Elements of the Measurement Covariance Matrix R.................. 234
8.3.2 Elements of the System State Covariance Matrix Q .................. 234
8.3.3 Initial Values of the Error Covariance Matrix P 0 ....................... 235

8.4 First Recursive Trials and Monitoring of Convergence ........................ 236
8.4.1 Jacobian Elements and Hypothesis Checking............................ 237
8.4.2 Monitoring Residues.................................................................. 241
8.5 Road Elements To Be Initialized........................................................... 241

8.6 Exploiting the Idea of Gestalt................................................................ 243
8.6.1 The Extended Gestalt Idea for Dynamic Machine Vision......... 245
8.6.2 Traffic Circle as an Example of Gestalt Perception ................... 251
8.7 Default Procedure for Objects of Unknown Classes ............................. 251

9 Recursive Estimation of Road Parameters
and Ego State While Cruising.......................................253

9.1 Planar Roads with Minor Perturbations in Pitch ................................... 255
9.1.1 Discrete Models ......................................................................... 255
9.1.2 Elements of the Jacobian Matrix................................................ 256
9.1.3 Data Fusion by Recursive Estimation ........................................ 257

9.1.4 Experimental Results ................................................................. 258
9.2 Hilly Terrain, 3-D Road Recognition.................................................... 259
9.2.1 Superposition of Differential Geometry Models........................ 260
9.2.2 Vertical Mapping Geometry....................................................... 261
9.2.3 The Overall 3-D Perception Model for Roads .......................... 262
9.2.4 Experimental Results ................................................................. 263

9.3 Perturbations in Pitch and Changing Lane Widths................................ 268
9.3.1 Mapping of Lane Width and Pitch Angle .................................. 268
9.3.2 Ambiguity of Road Width in 3-D Interpretation........................ 270