Automated Driving The Role of Forecasts and Uncertainty - A Control Perspective

Published on Jul 1, 2015in European Journal of Control1.54
· DOI :10.1016/J.EJCON.2015.04.007
Ashwin Carvalho15
Estimated H-index: 15
(University of California, Berkeley),
Stephanie Lefevre15
Estimated H-index: 15
(University of California, Berkeley)
+ 2 AuthorsFrancesco Borrelli69
Estimated H-index: 69
(University of California, Berkeley)
Sources
Abstract
Abstract Driving requires forecasts. Forecasted movements of objects in the driving scene are uncertain. Inevitably, decision and control algorithms for autonomous driving need to cope with such uncertain forecasts. In assisted driving, the uncertainty in the human/vehicle interaction further increases the complexity of the control design task. Our research over the past ten years has focused on control design methods which systematically handle uncertain forecasts for autonomous and semi-autonomous vehicles. This paper presents an overview of our findings and discusses relevant aspects of our recent results.
Download
📖 Papers frequently viewed together
522 Citations
2016
5 Authors (Brian Paden, ..., Emilio Frazzoli)
882 Citations
786 Citations
References74
Newest
#1Jason J. Kong (University of California, Berkeley)H-Index: 7
#2Mark Pfeiffer (ETH Zurich)H-Index: 9
Last. Francesco Borrelli (University of California, Berkeley)H-Index: 69
view all 4 authors...
We study the use of kinematic and dynamic vehicle models for model-based control design used in autonomous driving. In particular, we analyze the statistics of the forecast error of these two models by using experimental data. In addition, we study the effect of discretization on forecast error. We use the results of the first part to motivate the design of a controller for an autonomous vehicle using model predictive control (MPC) and a simple kinematic bicycle model. The proposed approach is l...
287 CitationsSource
#1Stephanie Lefevre (University of California, Berkeley)H-Index: 15
#2Ashwin Carvalho (University of California, Berkeley)H-Index: 15
Last. Francesco Borrelli (University of California, Berkeley)H-Index: 69
view all 3 authors...
We propose a learning-based method for the longitudinal control of an autonomous vehicle on the highway. We use a driver model to generate acceleration inputs which are used as a reference by a model predictive controller. The driver model is trained using real driving data, so that it can reproduce the driver's behavior. We show the system's ability to reproduce different driving styles from different drivers. By solving a constrained optimization problem, the model predictive controller ensure...
46 CitationsSource
#1Georg Schildbach (University of California, Berkeley)H-Index: 15
#2Francesco Borrelli (University of California, Berkeley)H-Index: 69
This paper presents a new algorithm for detecting the safety of lane changes on highways and for computing safe lane change trajectories. This task is considered as a building block for driver assistance systems and autonomous cars. The presented algorithm is based on recent results in Scenario Model Predictive Control (SCMPC). It accounts for the uncertainty in the traffic environment via a small number of future scenarios, which can be generated by any model-based or data-based approach. The p...
57 CitationsSource
#1Sanghyun Hong (University of California, Berkeley)H-Index: 7
#2Chankyu Lee (University of California, Berkeley)H-Index: 3
Last. J. Karl Hedrick (University of California, Berkeley)H-Index: 33
view all 4 authors...
This paper proposes a novel algorithm to identify three inertial parameters: sprung mass, yaw moment of inertia, and longitudinal position of the center of gravity. A four-wheel nonlinear vehicle model with roll dynamics and a correlation between the inertial parameters is used for a dual unscented Kalman filter to simultaneously identify the inertial parameters and the vehicle state. A local observability analysis on the nonlinear vehicle model is used to activate and deactivate different modes...
58 CitationsSource
#1Georg Schildbach (ETH Zurich)H-Index: 15
#2Lorenzo FagianoH-Index: 29
Last. Manfred Morari (ETH Zurich)H-Index: 123
view all 4 authors...
Many practical applications in control require that constraints on the inputs and states of the system are respected, while some performance criterion is optimized. In the presence of model uncertainties or disturbances, it is often sufficient to satisfy the state constraints for at least a prescribed share of the time, such as in building climate control or load mitigation for wind turbines. For such systems, this paper presents a new method of Scenario-Based Model Predictive Control (SCMPC). T...
161 CitationsSource
#1Stephanie Lefevre (University of California, Berkeley)H-Index: 15
#2Dizan Vasquez (IRIA: French Institute for Research in Computer Science and Automation)H-Index: 15
Last. Christian Laugier (IRIA: French Institute for Research in Computer Science and Automation)H-Index: 23
view all 3 authors...
With the objective to improve road safety, the automotive industry is moving toward more “intelligent” vehicles. One of the major challenges is to detect dangerous situations and react accordingly in order to avoid or mitigate accidents. This requires predicting the likely evolution of the current traffic situation, and assessing how dangerous that future situation might be. This paper is a survey of existing methods for motion prediction and risk assessment for intelligent vehicles. The propose...
522 CitationsSource
#1Victor Shia (University of California, Berkeley)H-Index: 11
#2Yiqi Gao (University of California, Berkeley)H-Index: 15
Last. Ruzena Bajcsy (University of California, Berkeley)H-Index: 75
view all 7 authors...
Threat assessment during semiautonomous driving is used to determine when correcting a driver's input is required. Since current semiautonomous systems perform threat assessment by predicting a vehicle's future state while treating the driver's input as a disturbance, autonomous controller intervention is limited to a restricted regime. Improving vehicle safety demands threat assessment that occurs over longer prediction horizons wherein a driver cannot be treated as a malicious agent. In this p...
95 CitationsSource
#1Stephanie Lefevre (University of California, Berkeley)H-Index: 15
#2Chao Sun (University of California, Berkeley)H-Index: 1
Last. Christian Laugier (IRIA: French Institute for Research in Computer Science and Automation)H-Index: 51
view all 4 authors...
Predicting the future speed of the ego-vehicle is a necessary component of many Intelligent Transportation Systems (ITS) applications, in particular for safety and energy management systems. In the last four decades many parametric speed prediction models have been proposed, the most advanced ones being developed for use in traffic simulators. More recently non-parametric approaches have been applied to closely related problems in robotics. This paper presents a comparative evaluation of paramet...
70 CitationsSource
#1Ali Mesbah (MIT: Massachusetts Institute of Technology)H-Index: 23
#2Stefan Streif (MIT: Massachusetts Institute of Technology)H-Index: 16
Last. Richard D. Braatz (MIT: Massachusetts Institute of Technology)H-Index: 75
view all 4 authors...
Stochastic uncertainties are ubiquitous in complex dynamical systems and can lead to undesired variability of system outputs and, therefore, a notable degradation of closed- loop performance. This paper investigates model predictive control of nonlinear dynamical systems subject to probabilistic parametric uncertainties. A nonlinear model predictive control framework is presented for control of the probability dis- tribution of system states while ensuring the satisfaction of constraints with so...
145 CitationsSource
#1Yiqi Gao (University of California, Berkeley)H-Index: 15
#2Andrew Gray (University of California, Berkeley)H-Index: 10
Last. Francesco Borrelli (University of California, Berkeley)H-Index: 69
view all 4 authors...
This paper proposes a robust control framework for lane-keeping and obstacle avoidance of semiautonomous ground vehicles. It presents a systematic way of enforcing robustness during the MPC design stage. A robust nonlinear model predictive controller (RNMPC) is used to help the driver navigating the vehicle in order to avoid obstacles and track the road centre line. A force-input nonlinear bicycle vehicle model is developed and used in the RNMPC control design. A robust invariant set is used in ...
113 CitationsSource
Cited By106
Newest
#1Heejin Ahn (UBC: University of British Columbia)
#1Heejin Ahn (UBC: University of British Columbia)H-Index: 7
Last. Ian M. Mitchell (UBC: University of British Columbia)H-Index: 27
view all 4 authors...
Source
#1Saeid SadeghiH-Index: 1
view all 0 authors...
Source
#1Zhixian Liu (Hunan University)H-Index: 1
#2Xiaofang Yuan (Hunan University)H-Index: 23
Last. Xizheng Zhang (HIE: Hunan Institute of Engineering)H-Index: 4
view all 5 authors...
Path planning is a basic function for autonomous vehicle (AV). However, it is difficult to adapt to different velocities and different types of obstacles including dynamic obstacle and static obstacle (such as road boundary) for AV. To solve the problem of path planning under different velocities and different types of obstacles, a two potential fields fused adaptive path planning system (TPFF-APPS) which includes two parts, a potential field fusion controller and an adaptive weight assignment u...
Source
#1Yunli Shao (ORNL: Oak Ridge National Laboratory)
#2Yuan Zheng (Seagate Technology)
Last. Zongxuan Sun (UMN: University of Minnesota)H-Index: 22
view all 3 authors...
Connected and autonomous vehicles (CAVs) can bring in energy, mobility, and safety benefits to transportation. The optimal control strategies of CAVs are usually determined for a look-ahead horizon using previewed traffic information. This requires the development of an effective future traffic prediction algorithm and its integration to the CAV control framework. However, it is challenging for short-term traffic prediction using information from connectivity, especially for mixed traffic scenar...
Source
#1Omveer Sharma (IIT BBS: Indian Institute of Technology Bhubaneswar)H-Index: 1
#2N C Sahoo (IIT BBS: Indian Institute of Technology Bhubaneswar)H-Index: 3
Last. Niladri B. Puhan (IIT BBS: Indian Institute of Technology Bhubaneswar)H-Index: 14
view all 3 authors...
Abstract Autonomous vehicles (AVs) have now drawn significant attentions in academic and industrial research because of various advantages such as safety improvement, lower energy and fuel consumption, exploitation of road network, reduced traffic congestion and greater mobility. In critical decision making process during motion of an AV, intelligent motion planning takes an important and challenging role for obstacle avoidance, searching for the safest path to follow, generation of suitable beh...
2 CitationsSource
#1Richard Cheng (California Institute of Technology)H-Index: 5
#2Richard M. Murray (California Institute of Technology)H-Index: 100
Last. Joel W. Burdick (California Institute of Technology)H-Index: 50
view all 3 authors...
When autonomous robots interact with humans, such as during autonomous driving, explicit safety guarantees are crucial in order to avoid potentially life-threatening accidents. Many data-driven methods have explored learning probabilistic bounds over human agents' trajectories (i.e. confidence tubes that contain trajectories with probability \delta, which can then be used to guarantee safety with probability 1-\delta However, almost all existing works consider \delta \geq 0.001 The purp...
1 Citations
#1Marcel MennerH-Index: 4
#2Karl BerntorpH-Index: 13
Last. Stefano Di CairanoH-Index: 21
view all 4 authors...
This paper presents a method for inverse learning of a control objective defined in terms of requirements and their joint probability distribution from data. The probability distribution characterizes tolerated deviations from the deterministic requirements and is learned using maximum likelihood estimation from data. Further, this paper introduces both parametrized requirements for motion planning in autonomous driving applications and methods for the estimation of their parameters from driving...
1 CitationsSource
In this article, the autonomous driving problem in an unstructured scene is addressed using a model predictive control (MPC) scheme. The lack of scene structure makes the sensing problem challenging, in particular when considered within a control loop. To circumvent this difficulty, the bundle adjustment (BA) algorithm from computer vision is used to detect obstacles and compute a sparse representation of the environment. In one of the main results of this article, it is shown how this sparse re...
3 CitationsSource
Dec 1, 2020 in GLOBECOM (Global Communications Conference)
#1Christian Vitale (UCY: University of Cyprus)H-Index: 5
#2Panayiotis Kolios (UCY: University of Cyprus)H-Index: 12
Last. Georgios Ellinas (UCY: University of Cyprus)H-Index: 27
view all 3 authors...
Intersections are among the most challenging sections of our road infrastructure and a clear bottleneck for traffic flows. Key aspects of the 5G cellular network, e.g., the Multi Access Edge Computational (MEC) platform and the reduced network latency, act as enablers for the utilization of Connected Autonomous Vehicles (CAVs) that can ultimately bring about drastic changes in the management of intersection crossings and transportation networks in general. To date, there exist extensive research...
Source
#2Mauricio MarcanoH-Index: 6
Last. Joshué PérezH-Index: 24
view all 5 authors...
Model-based trajectory tracking has become a widely used technique for automated driving system applications. A critical design decision is the proper selection of a vehicle model that achieves the best trade-off between real-time capability and robustness. Blending different types of vehicle models is a recent practice to increase the operating range of model-based trajectory tracking control applications. However, current approaches focus on the use of longitudinal speed as the blending parame...
3 CitationsSource