Study on Sensor Fusion Methodology for Vehicle Control to Handle Information on Driving Environment
Hiroshi Takahashi · TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C · 1997
This paper concerns the process of selecting control characteristics such as reference models and control patterns in vehicle control systems. Signals may be received from many different sensors, including a video camera, laser radar and other devices, in the process of executing vehicle control. With conventional control methods, it is not easy to integrate signals from so many sensors. The program routine becomes extremely complicated in terms of the procedural representation. A new method for integrating the signals from many sensors is proposed here with the aim of resolving this structural problem. This method is analogous to the analytic hierarchy process (AHP). In this approach, AHP is regarded as a mapping function from the space of sensor signals to the space of system variables such as evaluation values and control patterns. Declarative knowledge based on pairwise comparisons of two signals is prepared for use in integrating sensor information. The results of a simulation analysis are presented to validate the effectiveness of the proposed system in generating the control targets of a vehicle control system.