CAN-Bus signal analysis using stochastic methods and pattern recognition in time series for active safety

Abstract

In the development of driver-adaptive and context-aware active safety applications, CAN-Bus signals play a central role. Modern vehicles are equipped with several sensors and ECU (electronic control unit) to provide measurements for internal combustion engine and several active vehicle safety systems, such as ABS (anti-lock brake system) and ESP (electronic stability program). The entire communication between sensors, ECU, and actuators in a modern automobile is performed via the CAN-Bus. However, the long-term history and trends in the CAN-Bus signals, which contain important information on driving patterns and driver characteristics, has not been widely explored. The traditional engine and active safety systems use a very small time window (t < 2s) of the CAN-Bus to operate. On the contrary, the implementation of driver-adaptive and context-aware systems requires longer time windows and different methods for analysis. In this chapter, a summary of systems that can be built on this type of analysis is presented. The CAN-Bus signals are used to recognize the patterns in long-term representing driving subtasks, maneuvers, and routes. Based on the analysis results, quantitative metrics/feature vectors are suggested that can be used in many ways, with two prospects considered here: (1) CAN-Bus signals can be presented in a way to distinguish distracted/impaired driver behavior from normal/safe and (2) driver characteristics and control strategies can be quantitatively identified so that active safety controllers can be adapted accordingly to obtain the best driver-vehicle response for safe systems. In other words, an optimal human-machine cooperative system can be designed to achieve improved overall safety. © Springer Science+Business Media, LLC 2012.