Approximate Topological Optimization using Multi-Mode Estimation for Robot Motion Planning

In this extended abstract, we report on ongoing work towards an approximate multimodal optimization algorithm with asymptotic guarantees. Multimodal optimization is the problem of finding all local optimal solutions (modes) to a path optimization problem. This is important to compress path databases, as contingencies for replanning and as source of symbolic representations. Following ideas from Morse theory, we define modes as paths invariant under optimization of a cost functional. We develop a multi-mode estimation algorithm which approximately finds all modes of a given motion optimization problem and asymptotically converges. This is made possible by integrating sparse roadmaps with an existing single-mode optimization algorithm. Initial evaluation results show the multimode estimation algorithm as a promising direction to study path spaces from a topological point of view.

[1]  Torsten Bertram,et al.  Integrated online trajectory planning and optimization in distinctive topologies , 2017, Robotics Auton. Syst..

[2]  Andreas Orthey,et al.  Visualizing Local Minima in Multi-Robot Motion Planning using Multilevel Morse Theory. , 2020 .

[3]  Marc Toussaint,et al.  Bayesian Functional Optimization , 2018, AAAI.

[4]  Siddhartha S. Srinivasa,et al.  Implicit Multiagent Coordination at Unsignalized Intersections via Multimodal Inference Enabled by Topological Braids , 2020 .

[5]  Byron Boots,et al.  Functional Gradient Motion Planning in Reproducing Kernel Hilbert Spaces , 2016, Robotics: Science and Systems.

[6]  Takayuki Osa.  Multimodal trajectory optimization for motion planning , 2020, Int. J. Robotics Res..

[7]  Joshua A. Levine,et al.  The Topology ToolKit , 2018, IEEE Transactions on Visualization and Computer Graphics.

[8]  Marc Toussaint,et al.  Robot trajectory optimization using approximate inference , 2009, ICML '09.

[9]  Vijay Kumar,et al.  Topological constraints in search-based robot path planning , 2012, Auton. Robots.

[10]  Subhrajit Bhattacharya,et al.  Path homotopy invariants and their application to optimal trajectory planning , 2017, Annals of Mathematics and Artificial Intelligence.

[11]  Sébastien Vérel,et al.  Local Optima Networks: A New Model of Combinatorial Fitness Landscapes , 2014, ArXiv.

[12]  Ross A. Knepper,et al.  Multi-agent path topology in support of socially competent navigation planning , 2018, Int. J. Robotics Res..

[13]  Lydia E. Kavraki,et al.  The Open Motion Planning Library , 2012, IEEE Robotics & Automation Magazine.

[14]  Pieter Abbeel,et al.  Motion planning with sequential convex optimization and convex collision checking , 2014, Int. J. Robotics Res..

[15]  Oliver Brock,et al.  Contingent Contact-Based Motion Planning , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[16]  Barbora Kozlíková,et al.  Searching Multiple Approximate Solutions in Configuration Space to Guide Sampling-Based Motion Planning , 2020, J. Intell. Robotic Syst..

[17]  Hlawka.  The calculus of variations in the large , 1939 .

[18]  Byron Boots,et al.  Continuous-time Gaussian process motion planning via probabilistic inference , 2017, Int. J. Robotics Res..

[19]  Kostas E. Bekris,et al.  Sparse roadmap spanners for asymptotically near-optimal motion planning , 2014, Int. J. Robotics Res..

[20]  Marc Toussaint,et al.  Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning , 2018, Robotics: Science and Systems.

[21]  Stefan Schaal,et al.  STOMP: Stochastic trajectory optimization for motion planning , 2011, 2011 IEEE International Conference on Robotics and Automation.

[22]  Oliver Brock,et al.  Elastic roadmaps—motion generation for autonomous mobile manipulation , 2010, Auton. Robots.

[23]  Andreas Orthey,et al.  Multilevel Motion Planning: A Fiber Bundle Formulation , 2020, ArXiv.

[24]  Gabriela Ochoa,et al.  Modelling parameter configuration spaces with local optima networks , 2020, GECCO.

[25]  J F Williamson.  Random selection of points distributed on curved surfaces. , 1987, Physics in medicine and biology.

[26]  Mark H. Overmars,et al.  Multilevel Path Planning for Nonholonomic Robots Using Semiholonomic Subsystems , 1998, Int. J. Robotics Res..

[27]  Florian T. Pokorny,et al.  Topological trajectory classification with filtrations of simplicial complexes and persistent homology , 2016, Int. J. Robotics Res..

[28]  Thierry Siméon,et al.  Visibility-based probabilistic roadmaps for motion planning , 2000, Adv. Robotics.

[29]  Mike Preuss,et al.  Multimodal Optimization by Means of Evolutionary Algorithms , 2015, Natural Computing Series.

[30]  Marston Morse.  The Calculus of Variations in the Large , 1934 .

[31]  Danica Kragic,et al.  High-dimensional Winding-Augmented Motion Planning with 2D topological task projections and persistent homology , 2016, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[32]  Marc Toussaint,et al.  Newton methods for k-order Markov Constrained Motion Problems , 2014, ArXiv.

[33]  Marc Toussaint,et al.  Motion Planning Explorer: Visualizing Local Minima Using a Local-Minima Tree , 2020, IEEE Robotics and Automation Letters.

[34]  Thierry Siméon,et al.  Path Deformation Roadmaps: Compact Graphs with Useful Cycles for Motion Planning , 2008, Int. J. Robotics Res..

[35]  Ross A. Knepper,et al.  Decentralized Multi-Agent Navigation Planning with Braids , 2016, WAFR.