A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles

By Brian A. Paden, Michal Cáp, Sze Zheng Yong, Dmitry S. Yershov, Emilio Frazzoli. 2016

2,380 citations recorded by the source corpus.

Self-driving vehicles are a maturing technology with the potential to reshape mobility by enhancing the safety, accessibility, efficiency, and convenience of automotive transportation. Safety-critical tasks that must be executed by a self-driving vehicle include planning of motions through a dynamic environment shared with other vehicles and pedestrians, and their robust executions via feedback control. The objective of this paper is to survey the current state of the art on planning and control algorithms with particular regard to the urban setting. A selection of proposed techniques is reviewed along with a discussion of their effectiveness. The surveyed approaches differ in the vehicle mobility model used, in assumptions on the structure of the environment, and in computational requirements. The side by side comparison presented in this survey helps to gain insight into the strengths and limitations of the reviewed approaches and assists with system level design choices.

Subfields: Automotive Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Automotive Engineering

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Deep Reinforcement Learning for Autonomous Driving: A Survey (2020, neutral): “While the family of MPC methods aim to stabilize the behavior of the vehicle while tracking the speciﬁed path [14].”
Driving Policy Transfer via Modularity and Abstraction (2018, background): “ not validate their ideas with actual driving. Driving policies. Autonomous driving systems are commonly implemented via modular pipelines that comprise a multitude of carefully engineered components [14,26,33,49]. The advantage of this modular design is that each component can be developed in isola”
A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, background): “[20] X X X X X [21] X X X [22] X X [23] X [24] X X X X [25] X X X [26] X [27] X X X X X [28] X X X X Ours X X X X X X X X X”
End-to-End Driving Via Conditional Imitation Learning (2017, background): “Autonomous driving is the subject of intensive research [25].”
Agile Autonomous Driving using End-to-End Deep Imitation Learning (2017, background): “Consequently, designing a control policy by following the traditional model-plan-then-act approach [18, 21] becomes challenging, as it is difﬁcult to adequately characterize the robot’s interaction with the environment a priori .”
CARLA: An Open Urban Driving Simulator (2017, background): “Its most difﬁcult form is navigation in densely populated urban environments [21].”
Textual Explanations for Self-Driving Vehicles (2018, background): “The mediated perceptionbased approach depends on recognizing human-designated features, such as lane markings, traffic lights, pedestrians or cars, which generally require demanding parameter tuning for a balanced performance [19].”
Control of Connected and Automated Vehicles: State of the Art and Future Challenges (2018, methodology): “As a consequence, estimation, prediction, and learning techniques play a major role in this ﬁeld [59].”
Research Advances and Challenges of Autonomous and Connected Ground Vehicles (2019, methodology): “…utilizes the center of the lane or a coarse geometric path as input to search for the best driving options (behavior) (e.g. driving along a lane or turning at an intersection) in terms of the perceived behavior of other trafﬁc participants, road conditions, and signals from infrastructure [344].”
DLow: Diversifying Latent Flows for Diverse Human Motion Prediction (2020, background): “, predicting the future 3D poses of a person based on past poses, is an important problem in computer vision and has many useful applications in autonomous driving [55], human robot interaction [39] and healthcare [67].”
Computer-aided design for safe autonomous vehicles (2017, methodology): “The Trajectory Tracker (TT) computes acceleration and steering rate to make the vehicle follow the reference trajectory computed by the TP using the dynamics in [18].”
Self-Driving Cars: A Survey (2019, methodology): “Methods for motion planning can be mainly categorized into four classes: graph search based, sampling based, interpolating curve based, and numerical optimization based methods (González et al., 2015; Paden et al., 2016).”
Agile Off-Road Autonomous Driving Using End-to-End Deep Imitation Learning (2017, background): “Consequently, designing a control policy by following the traditional model-plan-then-act approach [1], [4] becomes challenging, as it is difﬁcult to adequately characterize the robot’s interaction with the environment a priori .”
A Review of Motion Planning for Highway Autonomous Driving (2020, background): “Unlike the classification in [25], we do not focus on the usual points of comparison, such as completeness, optimality, or time complexity, but on how the algorithms offer an effective and efficiently implementable response to practical”
Multimodal End-to-End Autonomous Driving (2019, background): “defined modules with particular responsibilities; for instance, to provide environmental perception [1], [2], as well as route planning and maneuver control [3], [4].”
Survey of Deep Reinforcement Learning for Motion Planning of Autonomous Vehicles (2020, background): “This decision-making system of autonomous driving can be decomposed into at least four layers, as stated in [3] (see Fig.2.).”