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

Paden, B.; Čáp, Michal; Yong, Sze Zheng; Yershov, Dmitry; Frazzoli, Emilio · 2016 · OpenAlex-citations

DOI: 10.1109/tiv.2016.2578706

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper surveys the state-of-the-art algorithms for motion planning and feedback control in self-driving vehicles, specifically within urban environments. Motivated by the potential of autonomous vehicles to reduce traffic fatalities—94% of which are attributed to driver error—and improve mobility accessibility, the authors focus on safety-critical tasks for automation levels 3 and above. The study excludes perception systems, instead concentrating on the decision-making hierarchy that processes environmental data to generate safe vehicle motions. The authors structure the decision-making process into four hierarchical components: route planning, behavioral decision making, motion planning, and vehicle control. Route planning utilizes graph-based algorithms on road networks to find minimum-cost paths. Behavioral decision making employs finite state machines or probabilistic frameworks, such as Markov Decision Processes (MDPs) and Partially Observable MDPs, to handle interactions with other traffic participants and uncertainty in their intentions. Motion planning translates behavioral decisions into dynamically feasible, collision-free paths or trajectories. Finally, vehicle control executes these plans using feedback controllers to correct tracking errors. The survey reviews specific modeling and algorithmic techniques for these components. For vehicle modeling, the paper details the kinematic single-track model, which approximates vehicle mobility using nonholonomic constraints, and discusses variations like the Dubins and Reeds-Shepp cars. It also addresses inertial effects for higher-fidelity modeling. In motion planning, the authors categorize methods into variational approaches (non-linear optimization), graph-search methods (including lane graphs, geometric methods, and sampling-based techniques like RRTs), and incremental search strategies. For control, the paper examines path stabilization methods such as Pure Pursuit and rear/front wheel position-based feedback, as well as trajectory tracking controllers including Control Lyapunov functions, output feedback linearization, Model Predictive Control (MPC), and Linear Parameter Varying controllers. The significance of this work lies in its comprehensive side-by-side comparison of these techniques, highlighting their strengths, limitations, and computational requirements. By analyzing how different approaches vary in their mobility models, environmental assumptions, and complexity, the survey provides critical insights for system-level design choices. It serves as a foundational reference for researchers and engineers developing autonomous driving systems, clarifying the trade-offs between model accuracy and computational tractability in real-time urban driving scenarios.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• situational awareness