Safe semi-autonomous control with enhanced driver modeling

Vasudevan, Ram; Shia, Victor; Gao, Yiqi; Cervera-Navarro, Ricardo; Bajcsy, Růžena; Borrelli, Francesco

doi:10.1109/acc.2012.6315654

Cited by 63 publications

(43 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned in [9], the driver's steering behavior is the inherent uncertainty for this system. Although some very accurate driver model estimation approaches are proposed, see [10] and [11], we use a simple uncertain driver model, as proposed in [12] in this paper, due to its simplicity in computation. The existing MPC controllers that do consider the disturbance of the driver, such as the ones in [13] and [14], are based on linearized dynamics of the vehicle, which is not applicable in all scenarios, especially when the slip angles of the tires are high.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear stochastic predictive control with unscented transformation for semi-autonomous vehicles

Liu

Gray

Lee

et al. 2014

2014 American Control Conference

View full text Add to dashboard Cite

This paper presents a novel predictive control approach based on the unscented transformation with recursive feasibility analysis and an experimental validation for lane keeping of semi-autonomous vehicles. The optimization problem to be solved is nonlinear with stochastic disturbances and probability constraints on states. The unscented transformation is utilized to calculate the propagation of disturbed states over the prediction horizon, and the probability constraints are transformed into constraint functions with Chebyshev's inequality. A sufficient condition for recursive feasibility is proved by considering the worst case of the disturbance realization. Experiments on the lane keeping system with an uncertain driver model validate the effectiveness of the proposed approach.

show abstract

Section: Introductionmentioning

confidence: 99%

Nonlinear stochastic predictive control with unscented transformation for semi-autonomous vehicles

Liu

Gray

Lee

et al. 2014

2014 American Control Conference

View full text Add to dashboard Cite

show abstract

“…To assure safety of the driver, as well as driving experience, researchers fo cus on threat assessment, decision making, path planning and control strategies, which are the main elements of ac tive safety system [1]. A large variety of literatures can easily be found in each aspect [1][2][3][4].…”

Section: Introductionmentioning

confidence: 99%

Robust semi-autonomous vehicle control for roadway departure and obstacle avoidance

Liu

Zheng

Pan

2013

2013 13th International Conference on Control, Automation and Systems (ICCAS 2013)

View full text Add to dashboard Cite

This paper presents the design of a robust active safety system for prevention of unintended collisions and roadway departures. The safety system shares control with the driver and corrects the driver's steering and braking action only if there is a potential risk. We formulate the safety problem as constrained optimization problem using uncertain evolution set based MPC. In the scheme, we compute the set of possible trajectories with uncertain driver model and issue minimal steering and (or) braking action, only if necessary, to keep the set in safety constraints. Human in loop simulation experiments were conducted, the results demonstrate the robustness of proposed controller to uncertain input from drivers.

show abstract

“…Regardless of where researchers stand on the spectrum, there is a common issue that needs to be addressed: driver acceptance and adaptability to smart cars. With warning systems, it is of concern whether drivers find such system useful or distractive; with semi-autonomous systems, predictions of driver behavior are used to decide whether control interventions are necessary [8]; and with autonomous systems, there is a push towards making the vehicle behave closely to human drivers so the sense of taking over is less intrusive [9]. Therefore, it is important to formalize a systematic way of modeling the human driver.…”

Section: Introductionmentioning

confidence: 99%

Modeling driver behavior during complex maneuvers

Lin

Tseng

Borrelli

2013

2013 American Control Conference

View full text Add to dashboard Cite

This paper focuses on the modeling of driver behaviors during maneuvers that involve tire saturation. Drifting maneuvers are performed by an expert driver on low friction surfaces. Driving data is recorded and used to aid the construction of two different driver models. The first model parallels the driver to a model predictive controller and the second model treats the driver as a switched linear state feedback controller. This paper compares the two approaches and presents a framework where the driver model is integrated into an autonomous controller in the form of an optimization problem.

show abstract

Safe semi-autonomous control with enhanced driver modeling

Cited by 63 publications

References 18 publications

Nonlinear stochastic predictive control with unscented transformation for semi-autonomous vehicles

Nonlinear stochastic predictive control with unscented transformation for semi-autonomous vehicles

Robust semi-autonomous vehicle control for roadway departure and obstacle avoidance

Modeling driver behavior during complex maneuvers

Contact Info

Product

Resources

About