The Driving School System: Learning Basic Driving Skills From a Teacher in a Real Car

Abstract

To offer increased security and comfort, advanced driver-assistance systems (ADASs) should consider individual driving styles. Here, we present a system that learns a human's basic driving behavior and demonstrate its use as ADAS by issuing alerts when detecting inconsistent driving behavior. In contrast to much other work in this area, which is based on or obtained from simulation, our system is implemented as a multithreaded parallel central processing unit (CPU)/graphics processing unit (GPU) architecture in a real car and trained with real driving data to generate steering and acceleration control for road following. It also implements a method for detecting independently moving objects (IMOs) for spotting obstacles. Both learning and IMO detection algorithms are data driven and thus improve above the limitations of model-based approaches. The system's ability to imitate the teacher's behavior is analyzed on known and unknown streets, and results suggest its use for steering assistance but limit the use of the acceleration signal to curve negotiation. We propose that this ability to adapt to the driver can lead to better acceptance of ADAS, which is an important sales argument.Index Terms-Advanced individualized driver-assistance system, driving, imitation learning, independently moving object (IMO), real-time system. A DVANCED driver-assistance systems (ADASs) that adapt to the individual driver have high potential in the car industry since they can reduce the risk of accidents while providing a high degree of comfort. Conventional systems are based on a general moment-to-moment assessment of road and driving parameters. To arrive at a judgment of the current Manuscript

Toward human-like motion planning in urban environments

“…4) a ⇤ lon and d ⇤ lon from model M normal clearly show stronger linearity thanã ⇤ lon andd ⇤ lon from model M tight , which also reflects that the human driver tends to behave more unpredictably during high-curvature negotations. But overall, they all have rather low determination coefficient values, confirming the argument made in [16] that the longitudinal accelerations have a great variance depending on hard-to-access variables like driver mood.…”

“…Only a few works have concerned human driving patterns, for example: [15] investigated the proper speed model that explains a driver's modulation of speed when entering or exiting a curve. [16] attempted to identify abnormal on-road driving behavior in freeway situations by learning a steering-speed mapping. Both works assumes overly simplified models that were not expressive in characterizing general curve negotiation, especially for high-curvature turnings.…”

Investigation on the evolution of a robotic controller for autonomous vehicle navigation

“…Markelic and collaborators [10] proposed a system to learn driving skills based on a human teacher. Driving School (DRI-VSCO) is implemented as a multi-threaded, parallel CPU-GPU architecture in a real car and it was trained with real driving data to produce steering and acceleration control for road following.…”

Vision-Based Autonomous Navigation Using Supervised Learning Techniques

“…A more recent work, Markelic et al [9], proposes a system that learns driving skills based on a human teacher. Driving School (DRIVSCO) is implemented as a multi-threaded, parallel CPU/GPU architecture in a real car and trained with real driving data to generate steering and acceleration control for road following.…”