Learning to drive the human way: a step towards intelligent vehicles

Pasquier, M.; Oentaryo, Richard J.

doi:10.1504/ijvas.2008.016477

Cited by 17 publications

(19 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The enhancement and novel feature presented in detail in this paper is the realization of Intelligent Speed Adaptation and Steering Control (ISASC) that allows the vehicle to anticipate and negotiate curves safely. This is a significant improvement over simple longitudinal and lateral control [5] [17], similar to that in Highways Automated Systems for instance, and also over our previous results in Intelligent Speed Adaptation (ISA) [21], which used anticipation for longitudinal control only. This paper is organized as follows: section II describes the main characteristics of the GenSoFNN-Yager employed in this work; section III presents the ISASC system for anticipating and negotiating curves; lastly section IV elaborates on the experiments conducted in simulation, the results obtained, and their analysis.…”

Section: Introductionsupporting

confidence: 66%

“…Since it is not viable to experiment directly with a real vehicle, a driving simulator has been implemented and used in many projects [12][14] [17]. The simulator allows capturing from a human driver the necessary examples needed by the system to learn its driving skill.…”

Section: A Driving Simulator and Trainingmentioning

confidence: 99%

“…[12][13] [14][15] [16] Our approach consists of modeling human driving expertise as a continuous decision-making process using a set of rules that relate sensory input to control output. Since designing such an expert rule base is difficult, we devise systems that can learn by example by observing the human driver [12] [17]. The learning component is a fuzzy neural system such as POPFNN [18], GenSoFNN [19], or the latest in our CI toolbox, the GenSoFNN-Yager i.e., Generic Self-Organizing Fuzzy Neural Network using the Yager inference scheme [20].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An intelligent driving system for automatically anticipating and negotiating road curves

Chan

Partouche

Pasquier

2007

2007 IEEE/RSJ International Conference on Intelligent Robots and Systems

Self Cite

View full text Add to dashboard Cite

Human error is the main cause of the many road traffic accidents which every year take the lives of millions of people and injure many more. Driving safety is thus a major concern leading to research in autonomous driving systems. One project at the Centre for Computational Intelligence (C 2 i), NTU, aims at using fuzzy neural architectures such as GenSoFNN-Yager to realize intelligent driving i.e., to learn to autonomously park, make U-turns, drive, and even decide when to change lane, overtake, etc. This paper presents recent work on Intelligent Speed Adaptation and Steering Control (ISASC), a novel feature of which is the ability to anticipate the road profile and negotiate curves safely. The proposed system was developed and tested on a driving simulator. Experimental results show the robustness of the system in learning from example the desired human driving expertise and applying this knowledge to negotiate new unseen roads.

show abstract

Section: Introductionsupporting

confidence: 66%

Section: A Driving Simulator and Trainingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An intelligent driving system for automatically anticipating and negotiating road curves

Chan

Partouche

Pasquier

2007

2007 IEEE/RSJ International Conference on Intelligent Robots and Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Velocity control was handled by the driver. Further imitation learning work by Pasquier and Oentaryo describes the learning of several driving skills with a fuzzy neural network [21]. The algorithms for lane following were tested in simulation.…”

Section: State Of the Artmentioning

confidence: 99%

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

Markelić

Kjær-Nielsen

Pauwels

et al. 2011

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

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

show abstract

“…Our preliminary analysis shows that compared to the benchmark algorithm [1], the computational complexity for implementation of the proposed algorithm is indeed higher due mainly to the computations in MLFFNN. However, a number of experiments in literature has illustrated the feasibility of deploying neural networks in software [21,39].…”

Section: Appendix a Complexity Analysis For Implementationmentioning

confidence: 99%

Distributed Classification of Traffic Anomalies Using Microscopic Traffic Variables

Thajchayapong

Garcia-Trevino

Barria

2013

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

Abstract-This paper proposes a novel anomaly classification algorithm that can be deployed in a distributed manner and utilizes microscopic traffic variables shared by neighbouring vehicles to detect and classify traffic anomalies under different traffic conditions. The algorithm which incorporates multi-resolution concepts is based on the likelihood estimation of a neural network output and a bisection-based decision threshold. We show that when applied to real-world traffic scenarios, the proposed algorithm can detect all the traffic anomalies of the reference test data set; this represents a significant improvement over our previously proposed algorithm [1]. We also show that the proposed algorithm can effectively detect and classify traffic anomalies even when i) the microscopic traffic variables are available from only a fraction of the vehicle population and ii) some microscopic traffic variables are lost due to degradation in V2V and/or V2I communications.

show abstract

Learning to drive the human way: a step towards intelligent vehicles

Cited by 17 publications

References 24 publications

An intelligent driving system for automatically anticipating and negotiating road curves

An intelligent driving system for automatically anticipating and negotiating road curves

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

Distributed Classification of Traffic Anomalies Using Microscopic Traffic Variables

Contact Info

Product

Resources

About