Scalable Parallelization of Sampling-Based Planning Algorithms
We have developed a scalable method for parallelizing sampling-based planning algorithms. The method subdivides the configuration space (C-space) into regions and uses (sequential) sampling-based planners to build roadmaps in each region. The regional roadmaps are later connected to form a roadmap of the entire free space. By subdividing the space,we reduce the work and inter-processor communication associated with nearest neighbor calculations, a critical drawback and bottleneck to scalability in previous parallel motion planning methods.
There are two main types of sampling-based planning methods, graph-based methods used when multiple queries will be performed in the same environment (e.g., Probabilistic Roadmap Methods, or PRMs) and tree-based approached most suitable for single query settings (e.g., Rapidly-explorign Random Trees, or RRTs). We show how both types of methods can be parallelized efficiently.
Parallelizing Graph-based Methods
The multi-query graph-based methods (e.g., PRMs) build a graph during preprocessing which is later used to answer many queries. The constuction of the graph consists of two main phases, node generation and node connection, both of which can be easily parallelized. To generate n random nodes in parallel, we simply ask each of the p processors to generate n/p of them. Effective parallelization of the connection phase is essential to obtain scalable speedups since this typically accounts for 98% of the sequential computation time. A simple sequential connection algorithm attempts to connect every node to its k nearest neighbors. Suppose we group the n nodes into p sets, one for each processor. Then, a simple parallel solution is for each processor to compute the k nearest neighbors for its set of nodes and attempt the connections.
Parallelizing Tree-based Methods
The single-query tree-based methods (e.g., RRTs) have been more challenging to parallelize efficiently due to the sequential nature of the tree growth. In one approach, Radial RRT, we demonstrated a scalable algorithm that subdivides the space radially into regions to increase the computation locality. However, if an obstacle completely blocks RRT growth in a region, the planning space is not covered and the method is thus not probabilistically complete. We addressed this with another version, Blind RRT, which ignores obstacles during initial growth to efficiently explore the entire space. Because obstacles are ignored, free components of the tree become disconnected and fragmented. Blind RRT merges parts of the tree that have become disconnected from the root. We show how this algorithm can be applied to the Radial RRT framework allowing both scalability and effectiveness in motion planning. This method is a probabilistically complete approach to parallel RRTs. We show that our method not only scales but also overcomes the motion planning limitations that Radial RRT has in a series of difficult motion planning tasks.
| Distributed RRT Illustration |
|---|
![]() |
| Figure 1: Example of a radial subdivision for a 2D CSpace from a root node. Each process concurrently builds a subtree from the root and attempts to connect to goal. |
Results
We show that our method (pSBMP-X) is general enough to handle both graph-based (pSBMP-PRM) and tree-based (pSBMP-RRT) approaches. We compare our approach (pSBMP-PRM and pSBMP-RRT) to two other existing parallel algorithms (pPRM and pSRT) and demonstrate that our approach achieves better and more scalable performance (shown in Figure 1). Our approach achieves almost linear scalability to hundreds of cores (shown in Figure 2) on a 2400 cores Linux cluster of 13.8 tera-flop peak performance at Texas A&M University and over a thousand cores (shown in Figure 3) on a Cray XE6 petascale machine with peak performance of 1.288 peta-flops and 153,216 cores at Lawrence Berkeley National Laboratory.
Load Balancing Strategies
These strategies based on uniform spatial subdivision techniques for parallelizing sampling-based motion planning algorithms can scale well. However, such methods are prone to load imbalance, as planning time depends on region characteristics and, for most problems, the heterogeneity of the sub problems increases as the number of processors increases. We introduced two techniques to address load imbalance in the parallelization of sampling-based motion planning algorithms: an adaptive work stealing approach and bulk-synchronous redistribution. We proposed two techniques to address the problems of load imbalance. The first is a bulk-synchronous redistribution technique that redistributes regions among processors so that they have a more balanced distribution of data. The method approximates the amount of work a region will requre that is based on the complexity of a region and use it to attempt to balance work across processors, while also preserving the spatial geometry of the subdivision. The second approach is an adaptive work-stealing approach that permantently migrates both the region and the work related to it to improve the load balance.
Our results show that applying these techniques to representatives of the two major classes of parallel sampling-based motion planning algorithms, probabilistic roadmaps and rapidly-exploring random trees, results in a more scalable and load-balanced computation on more than 3,000 cores.
Publications
- Lee, H. , Motes, J. , Morales, M. , & Amato, N.M. (2021). Parallel Hierarchical Composition Conflict-Based Search. IEEE/RSJ International Conference on Intelligent Robots and Systems , 6(4) , 7001--7008. https://doi.org/10.1109/LRA.2021.3096476.
- Jacobs, S.A. & Amato, N.M. (2014). The anatomy of a distributed motion planning roadmap. 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems , 3019-3026. https://doi.org/10.1109/IROS.2014.6942979
- Jacobs, S.A. (2014). A Scalable Framework for Parallelizing Sampling-Based Motion Planning Algorithms. Doctoral Dissertation, Texas A&M University. View publication
- Fidel, A. , Jacobs, S.A. , Sharma, S. , Amato, N.M. , & Rauchwerger, L. (2014). Using Load Balancing to Scalably Parallelize Sampling-Based Motion Planning Algorithms. 2014 IEEE 28th International Parallel and Distributed Processing Symposium , 573-582. https://doi.org/10.1109/IPDPS.2014.66
- Rodriguez, C. , Denny, J. , Jacobs, S.A. , Thomas, S. , & Amato, N.M. (2013). Blind RRT: A Probabilistically Complete Distributed RRT. In. Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , 1758-1765. https://doi.org/10.1109/IROS.2013.6696587
- Ekenna, C. , Jacobs, S.A. , Thomas, S. , & Amato, N.M. (2013). Adaptive Neighbor Connection for PRMs: A Natural Fit for Heterogeneous Environments and Parallelism. In. Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , 1249-1256. https://doi.org/10.1109/IROS.2013.6696510
- Jacobs, S.A. , Stradford, N. , Rodriguez, C. , Thomas, S. , & Amato, N.M. (2013). A scalable distributed RRT for motion planning. 2013 IEEE International Conference on Robotics and Automation , 5088-5095. https://doi.org/10.1109/ICRA.2013.6631304
- Fidel, A. , Jacobs, S.A. , Sharma, S. , Rauchwerger, L. , & Amato, N.M. (2013). Load Balancing Techniques for Scalable Parallelization of Sampling-Based Motion Planning Algorithms. Technical Report, TR13-002 , Parasol Laboratory, Department of Computer Science, Texas A&M University.
- Jacobs, S.A. , Manavi, K. , Burgos, J. , Denny, J. , Thomas, S. , & Amato, N.M. (2012). A Scalable Method for Parallelizing Sampling-Based Motion Planning Algorithms. 2012 IEEE International Conference on Robotics and Automation , 2529-2536. https://doi.org/10.1109/ICRA.2012.6225334
- Jacobs, S.A. & Amato, N.M. (2011). From Days to Seconds: Scalable Parallel Algorithms for Motion Planning. In ACM Student Research Compet, Conf. on High Performance Computing Networking, Storage and Analysis Companion Proceedings. View publication
- Thomas, S. , Tanase, G. , Dale, L.K. , Moreira, J.E. , Rauchwerger, L. , & Amato, N.M. (2005). Parallel Protein Folding with STAPL. Concurrency and Computation: Practice and Experience , 17(14) , 1643-1656. https://doi.org/https://doi.org/10.1002/cpe.950
- Thomas, S. & Amato, N.M. (2004). Parallel Protein Folding with STAPL. 18th International Parallel and Distributed Processing Symposium (IPDPS) , 189-. https://doi.org/10.1109/IPDPS.2004.1303204
- Amato, N. & Dale, L. (1999). Probabilistic Roadmap Methods are Embarrassingly Parallel. Proceedings of IEEE International Conference on Robotics and Automation , 1 , 688-694 vol.1. https://doi.org/10.1109/ROBOT.1999.770055
