Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
158 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Design and Evaluation of Motion Planners for Quadrotors in Environments with Varying Complexities (2309.13720v2)

Published 24 Sep 2023 in cs.RO

Abstract: Motion planning techniques for quadrotors have advanced significantly over the past decade. Most successful planners have two stages: a front-end that determines a path that incorporates geometric (or kinematic or input) constraints and specifies the homotopy class of the trajectory, and a back-end that optimizes this path to respect dynamics and input constraints. While there are many different choices for each stage, the eventual performance depends critically not only on these choices, but also on the environment. Given a new environment, it is difficult to decide a priori how one should design a motion planner. In this work, we develop (i) a procedure to construct parametrized environments, (ii) metrics that characterize the difficulty of motion planning in these environments, and (iii) an open-source software stack that can be used to combine a wide variety of two-stage planners seamlessly. We perform experiments in simulations and a real platform. We find, somewhat conveniently, that geometric front-ends are sufficient for environments with varying complexities if combined with dynamics-aware backends. The metrics we designed faithfully capture the planning difficulty in a given environment. All code is available at https://github.com/KumarRobotics/kr_mp_design

Definition Search Book Streamline Icon: https://streamlinehq.com
References (42)
  1. P. Hart, N. Nilsson, and B. Raphael, “A formal basis for the heuristic determination of minimum cost paths,” IEEE Transactions on Systems Science and Cybernetics, vol. 4, no. 2, pp. 100–107, 1968.
  2. D. Harabor and A. Grastien, “Online graph pruning for pathfinding on grid maps,” in Proceedings of the Twenty-Fifth AAAI Conference on Artificial Intelligence, 2011, p. 1114–1119.
  3. S. M. LaValle, “Rapidly-Exploring Random Trees: A New Tool for Path Planning,” Tech. Rep., 1998.
  4. S. Liu, N. Atanasov, K. Mohta, and V. Kumar, “Search-based motion planning for quadrotors using linear quadratic minimum time control,” in 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017, pp. 2872–2879.
  5. Y. Li, Z. Littlefield, and K. E. Bekris, “Sparse methods for efficient asymptotically optimal kinodynamic planning,” in Algorithmic Foundations of Robotics XI: Selected Contributions of the Eleventh International Workshop on the Algorithmic Foundations of Robotics.   Springer, 2015, pp. 263–282.
  6. L. Jarin-Lipschitz, J. Paulos, R. Bjorkman, and V. Kumar, “Dispersion-minimizing motion primitives for search-based motion planning,” in 2021 IEEE International Conference on Robotics and Automation (ICRA), 2021, pp. 12 625–12 631.
  7. S. Teng, A. Jasour, R. Vasudevan, and M. G. Jadidi, “Convex Geometric Motion Planning on Lie Groups via Moment Relaxation,” in Proceedings of Robotics: Science and Systems, Daegu, Republic of Korea, July 2023.
  8. R. Tedrake, I. R. Manchester, M. Tobenkin, and J. W. Roberts, “Lqr-trees: Feedback motion planning via sums-of-squares verification,” The International Journal of Robotics Research, vol. 29, no. 8, pp. 1038–1052, 2010.
  9. T. Marcucci, J. Umenberger, P. A. Parrilo, and R. Tedrake, “Shortest paths in graphs of convex sets,” arXiv preprint arXiv:2101.11565, 2021.
  10. S. Liu, M. Watterson, K. Mohta, K. Sun, S. Bhattacharya, C. J. Taylor, and V. Kumar, “Planning dynamically feasible trajectories for quadrotors using safe flight corridors in 3-d complex environments,” IEEE Robotics and Automation Letters, pp. 1–1, 2017.
  11. B. Zhou, F. Gao, J. Pan, and S. Shen, “Robust Real-time UAV Replanning Using Guided Gradient-based Optimization and Topological Paths,” in Proceedings - IEEE International Conference on Robotics and Automation, 2020, pp. 1208–1214.
  12. E. Soria, F. Schiano, and D. Floreano, “Predictive control of aerial swarms in cluttered environments,” Nature Machine Intelligence, vol. 3, no. 6, pp. 545–554, 2021.
  13. J. Park, Y. Lee, I. Jang, and H. J. Kim, “Dlsc: Distributed multi-agent trajectory planning in maze-like dynamic environments using linear safe corridor,” IEEE Transactions on Robotics, pp. 1–20, 2023.
  14. L. Janson, E. Schmerling, A. Clark, and M. Pavone, “Fast marching tree: A fast marching sampling-based method for optimal motion planning in many dimensions,” The International journal of robotics research, vol. 34, no. 7, pp. 883–921, 2015.
  15. Y. Tao, Y. Wu, B. Li, F. Cladera, A. Zhou, D. Thakur, and V. Kumar, “Seer: Safe efficient exploration for aerial robots using learning to predict information gain,” in 2023 IEEE International Conference on Robotics and Automation (ICRA), 2023, pp. 1235–1241.
  16. K. Yang and S. Sukkarieh, “3d smooth path planning for a uav in cluttered natural environments,” in 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2008, pp. 794–800.
  17. D. Bareiss and J. van den Berg, “Reciprocal collision avoidance for robots with linear dynamics using lqr-obstacles,” in 2013 IEEE International Conference on Robotics and Automation, 2013, pp. 3847–3853.
  18. A. Loquercio, E. Kaufmann, R. Ranftl, M. Müller, V. Koltun, and D. Scaramuzza, “Learning high-speed flight in the wild,” Science Robotics, vol. 6, no. 59, p. eabg5810, 2021.
  19. R. Deits and R. Tedrake, “Efficient mixed-integer planning for uavs in cluttered environments,” in 2015 IEEE International Conference on Robotics and Automation (ICRA), 2015, pp. 42–49.
  20. H. Oleynikova, A. Millane, Z. Taylor, E. Galceran, J. Nieto, and R. Siegwart, “Signed distance fields: A natural representation for both mapping and planning,” in RSS 2016 workshop: geometry and beyond-representations, physics, and scene understanding for robotics.   University of Michigan, 2016.
  21. H. Oleynikova, Z. Taylor, M. Fehr, R. Siegwart, and J. Nieto, “Voxblox: Incremental 3d euclidean signed distance fields for on-board mav planning,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017.
  22. C. Chamzas, C. Quintero-Peña, Z. Kingston, A. Orthey, D. Rakita, M. Gleicher, M. Toussaint, and L. E. Kavraki, “Motionbenchmaker: A tool to generate and benchmark motion planning datasets,” IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 882–889, 2022.
  23. T. T. D. Montcel, A. Nègre, J.-E. Gomez-Balderas, and N. Marchand, “Boarr : A benchmark for quadrotor obstacle avoidance using ros and rotors,” ROSCon France, 2019.
  24. B. Boroujerdian, H. Genc, S. Krishnan, W. Cui, A. Faust, and V. Reddi, “Mavbench: Micro aerial vehicle benchmarking,” in 2018 51st annual IEEE/ACM international symposium on microarchitecture (MICRO), 2018, pp. 894–907.
  25. H. Yu, G. C. E. de Croon, and C. De Wagter, “Avoidbench: A high-fidelity vision-based obstacle avoidance benchmarking suite for multi-rotors,” in 2023 IEEE International Conference on Robotics and Automation (ICRA), 2023, pp. 9183–9189.
  26. W. Rehberg, J. Ortiz-Haro, M. Toussaint, and W. Hönig, “Comparison of optimization-based methods for energy-optimal quadrotor motion planning,” arXiv preprint arXiv:2304.14062, 2023.
  27. C. Nous, R. Meertens, C. De Wagter, and G. de Croon, “Performance evaluation in obstacle avoidance,” in 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2016, pp. 3614–3619.
  28. B. Cohen, I. A. Şucan, and S. Chitta, “A generic infrastructure for benchmarking motion planners,” in 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2012, pp. 589–595.
  29. M. Moll, I. A. Sucan, and L. E. Kavraki, “An extensible benchmarking infrastructure for motion planning algorithms,” arXiv preprint arXiv:1412.6673, 2014.
  30. J. B. Kruskal, “On the shortest spanning subtree of a graph and the traveling salesman problem,” Proceedings of the American Mathematical society, vol. 7, no. 1, pp. 48–50, 1956.
  31. M. Chen, Q. Hu, Z. Yu, H. THOMAS, A. Feng, Y. Hou, K. McCullough, F. Ren, and L. Soibelman, “Stpls3d: A large-scale synthetic and real aerial photogrammetry 3d point cloud dataset,” in 33rd British Machine Vision Conference (BMVC), London, UK, November 21-24, 2022. [Online]. Available: https://bmvc2022.mpi-inf.mpg.de/0429.pdf
  32. K. Chaney, F. Cladera, Z. Wang, A. Bisulco, M. A. Hsieh, C. Korpela, V. Kumar, C. J. Taylor, and K. Daniilidis, “M3ed: Multi-robot, multi-sensor, multi-environment event dataset,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June 2023, pp. 4015–4022.
  33. J. R. Munkres, “Topology; a first course.”   Prentice-Hall, 1974.
  34. S. Bhattacharya, “Search-based path planning with homotopy class constraints,” in Proceedings of the AAAI conference on artificial intelligence, vol. 24, no. 1, 2010, pp. 1230–1237.
  35. D. Mellinger and V. Kumar, “Minimum snap trajectory generation and control for quadrotors,” in 2011 IEEE International Conference on Robotics and Automation, 2011, pp. 2520–2525.
  36. F. Gao, W. Wu, J. Pan, B. Zhou, and S. Shen, “Optimal time allocation for quadrotor trajectory generation,” in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, pp. 4715–4722.
  37. Z. Wang, X. Zhou, C. Xu, and F. Gao, “Geometrically Constrained Trajectory Optimization for Multicopters,” IEEE Transactions on Robotics, vol. 38, no. 5, pp. 3259–3278, 2022.
  38. J. Tordesillas and J. P. How, “FASTER: Fast and safe trajectory planner for navigation in unknown environments,” IEEE Transactions on Robotics, 2021.
  39. S. Sun, A. Romero, P. Foehn, E. Kaufmann, and D. Scaramuzza, “A comparative study of nonlinear mpc and differential-flatness-based control for quadrotor agile flight,” IEEE Transactions on Robotics, vol. 38, no. 6, pp. 3357–3373, 2022.
  40. T. A. Howell, B. E. Jackson, and Z. Manchester, “Altro: A fast solver for constrained trajectory optimization,” in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019, pp. 7674–7679.
  41. I. A. Şucan, M. Moll, and L. E. Kavraki, “The Open Motion Planning Library,” IEEE Robotics & Automation Magazine, vol. 19, no. 4, pp. 72–82, December 2012, https://ompl.kavrakilab.org.
  42. M. Moll, I. A. Sucan, and L. E. Kavraki, “Benchmarking motion planning algorithms: An extensible infrastructure for analysis and visualization,” IEEE Robotics & Automation Magazine, vol. 22, no. 3, pp. 96–102, 2015.
Citations (3)

Summary

  • The paper introduces a modular two-stage planning framework that combines front-end trajectory generation with back-end optimization.
  • It quantifies environment complexity using the ECS metric, facilitating a systematic evaluation of planner performance.
  • Experiments reveal that RRT* paired with GCOPTER excels in high-density settings, while simpler scenarios may use alternative planners.

Design and Evaluation of Motion Planners for Quadrotors in Environments with Varying Complexities

The paper "Design and Evaluation of Motion Planners for Quadrotors in Environments with Varying Complexities" presented by Shao et al., explores the intricacies of motion planning for quadrotors, particularly in diverse and complex environments. It offers a systematic paper on how different planning algorithms can be evaluated and selected for optimal performance based on environmental factors.

Framework and Methodology

The authors propose a comprehensive framework for testing and evaluating motion planners across a spectrum of environments with varying degrees of complexity. The research focuses on a two-stage planning approach, where an initial trajectory is determined by a front-end planner, which is later refined by a back-end optimizer. This modular approach allows the evaluation of different combinations of front-end and back-end planners to determine the optimal configuration for a specific environmental complexity.

The paper also introduces the concept of an "Environment Complexity Signature" (ECS), which accounts for metrics such as density, clutter, and structure of the environment. These metrics provide a quantitative basis for understanding the difficulty level of environments and enable a systematic comparison of planner performance across these different settings.

Experimental Setup

Extensive simulations and experiments are conducted in both synthetic environments — such as randomly generated mazes and obstacle maps — and real-world scenarios. The ECS metrics guide these evaluations by providing a standardized description of the environment that is agnostic to the scale and orientation of the quadrotor, thus ensuring broader applicability of the results.

The experimental results highlight that geometric front-end planners (e.g., RRT*) combined with back-end trajectory optimizers (e.g., GCOPTER) display robust performance across a range of environments. Surprisingly, planners integrating dynamic constraints show negligible advantages in simpler settings and sometimes worse results in complex scenarios.

Key Insights and Findings

  • Modular Planning Approach: The two-stage hierarchy for motion planning allows for the decoupling of pathfinding and trajectory optimization, facilitating enhanced performance and adaptability to environment-specific challenges.
  • Importance of Environment Metrics: The ECS provides critical insights into how environmental complexity influences planner selection and performance, bolstering the notion of tailored planner design based on specific environmental configurations.
  • Optimal Planner Selection: The paper offers practical recommendations for selecting and designing motion planners. High-density environments benefit from the RRT* and GCOPTER combination, while in less cluttered conditions, alternatives like ALTRO and MPL prove advantageous in trajectory smoothness and computational efficiency.

Practical and Theoretical Implications

This research holds significant practical implications for the deployment of autonomous quadrotors in real-world applications such as urban navigation, delivery systems, and reconnaissance missions where environmental conditions can greatly vary. Theoretically, it pushes the envelope in understanding the intricate balance between planner complexity and environmental dynamics.

Future Directions

Future work could explore the integration of learning-based methods to dynamically adapt planner choice in real-time as environmental contexts evolve. Furthermore, extending the ECS framework to incorporate elements like environmental unpredictability or dynamic obstacles could enhance the robustness and reliability of motion planners in even more challenging scenarios.

In conclusion, the paper offers a well-structured and in-depth examination of motion planning for quadrotors, providing valuable insights and methodologies that are instrumental in advancing the field of autonomous aerial robotics.

Github Logo Streamline Icon: https://streamlinehq.com
Youtube Logo Streamline Icon: https://streamlinehq.com