Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Results of the 2024 CommonRoad Motion Planning Competition for Autonomous Vehicles

Yanliang Huang, Xia Yan, Peiran Yin, Zhenduo Zhang, Zeyan Shao, Youran Wang, Haoliang Huang, Matthias Althoff

TL;DR

The paper presents the 2024 CommonRoad Motion Planning Competition, benchmarking autonomous-vehicle planners on realistic, standardized scenarios with a consistent evaluation pipeline (feasibility plus a multi-term cost). It highlights two top approaches: a scalable, sampling-based Frenét-frame planner with level-k game-theoretic reasoning for intersections (TUM-2024) and an optimization-plus-reachability strategy (SB-2023). Results show TUM-2024 achieves substantially higher scenario coverage under a fixed compute budget, while SB-2023 delivers higher per-benchmark quality when both solvers succeed. The work underscores a trade-off between broad coverage and optimality, and points to improved prediction models as a path to closer performance between approaches.

Abstract

Over the past decade, a wide range of motion planning approaches for autonomous vehicles has been developed to handle increasingly complex traffic scenarios. However, these approaches are rarely compared on standardized benchmarks, limiting the assessment of relative strengths and weaknesses. To address this gap, we present the setup and results of the 4th CommonRoad Motion Planning Competition held in 2024, conducted using the CommonRoad benchmark suite. This annual competition provides an open-source and reproducible framework for benchmarking motion planning algorithms. The benchmark scenarios span highway and urban environments with diverse traffic participants, including passenger cars, buses, and bicycles. Planner performance is evaluated along four dimensions: efficiency, safety, comfort, and compliance with selected traffic rules. This report introduces the competition format and provides a comparison of representative high-performing planners from the 2023 and 2024 editions.

Results of the 2024 CommonRoad Motion Planning Competition for Autonomous Vehicles

TL;DR

The paper presents the 2024 CommonRoad Motion Planning Competition, benchmarking autonomous-vehicle planners on realistic, standardized scenarios with a consistent evaluation pipeline (feasibility plus a multi-term cost). It highlights two top approaches: a scalable, sampling-based Frenét-frame planner with level-k game-theoretic reasoning for intersections (TUM-2024) and an optimization-plus-reachability strategy (SB-2023). Results show TUM-2024 achieves substantially higher scenario coverage under a fixed compute budget, while SB-2023 delivers higher per-benchmark quality when both solvers succeed. The work underscores a trade-off between broad coverage and optimality, and points to improved prediction models as a path to closer performance between approaches.

Abstract

Over the past decade, a wide range of motion planning approaches for autonomous vehicles has been developed to handle increasingly complex traffic scenarios. However, these approaches are rarely compared on standardized benchmarks, limiting the assessment of relative strengths and weaknesses. To address this gap, we present the setup and results of the 4th CommonRoad Motion Planning Competition held in 2024, conducted using the CommonRoad benchmark suite. This annual competition provides an open-source and reproducible framework for benchmarking motion planning algorithms. The benchmark scenarios span highway and urban environments with diverse traffic participants, including passenger cars, buses, and bicycles. Planner performance is evaluated along four dimensions: efficiency, safety, comfort, and compliance with selected traffic rules. This report introduces the competition format and provides a comparison of representative high-performing planners from the 2023 and 2024 editions.
Paper Structure (14 sections, 8 equations, 3 figures, 3 tables)

This paper contains 14 sections, 8 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Format of the Competition
  3. Evaluation
  4. Feasible Trajectories
  5. Cost Function
  6. Winners
  7. Prediction Model
  8. Modified Reactive Planner
  9. Level-k Game Theory Planner for Intersections
  10. Results
  11. Evaluation Setup
  12. Benchmark Coverage Comparison
  13. Map-wise Performance Comparison
  14. Conclusion

Figures (3)

  • Figure 1: Example for a motion planning problem defined by the CommonRoad scenario DEU_Flensburg-73_1_T-1, where the goal set is depicted in yellow, the initial state of the ego vehicle in green, and the other traffic participants in blue.
  • Figure 2: Workflow of the motion planner.
  • Figure 3: Scenario coverage of the two motion planners: TUM-2024 and SB-2023.