Optimized Directed Roadmap Graph for Multi-Agent Path Finding Using Stochastic Gradient Descent

To read the file of this research, you can request a copy directly from the authors.


We present a novel approach called Optimized Directed Roadmap Graph (ODRM). It is a method to build a directed roadmap graph that allows for collision avoidance in multi-robot navigation. This is a highly relevant problem, for example for industrial autonomous guided vehicles. The core idea of ODRM is, that a directed roadmap can encode inherent properties of the environment which are useful when agents have to avoid each other in that same environment. Like Probabilistic Roadmaps (PRMs), ODRM's first step is generating samples from C-space. In a second step, ODRM optimizes vertex positions and edge directions by Stochastic Gradient Descent (SGD). This leads to emergent properties like edges parallel to walls and patterns similar to two-lane streets or roundabouts. Agents can then navigate on this graph by searching their path independently and solving occurring agent-agent collisions at run-time. Using the graphs generated by ODRM compared to a non-optimized graph significantly fewer agent-agent collisions happen. We evaluate our roadmap with both, centralized and decentralized planners. Our experiments show that with ODRM even a simple centralized planner can solve problems with high numbers of agents that other multi-agent planners can not solve. Additionally, we use simulated robots with decentralized planners and online collision avoidance to show how agents are a lot faster on our roadmap than on standard grid maps.

No file available

Request Full-text Paper PDF

To read the file of this research,
you can request a copy directly from the authors.

ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
In this paper we address the problem of online tra-jectory optimization and cooperative collision avoidance when multiple mobile service robots are operating in close proximity to each other. Using cooperative trajectory optimization to obtain smooth transitions in multi-agent path crossing scenarios applies to the demand for more flexibility and eciency in industrial autonomous guided vehicle (AGV) systems. We introduce a general approach for online trajectory optimization in dynamic environments. It involves an elastic-band based method for time-dependent obstacle avoidance combined with a model predictive trajectory planner that takes into account the robot's kinematic and kinodynamic constraints. We augment that planning approach to be able to cope with shared trajectories of other agents and perform an potential field based cooperative trajectory optimization. Performance and practical feasibility of the proposed approach are demonstrated in simulation as well as in real world experiments carried out on a representative set of path crossing scenarios with two industrial mobile service robots.
Conference Paper
Full-text available
This work proposes a motion planning method based on the construction of a roadmap connecting the critical points of a potential field or a distance function. It aims to overcome the limitation of potential field methods due to local minima caused by concave obstacles. The roadmap is incrementally constructed by a two-step procedure. Starting from a minimum, adjacent saddle-points are found using a local saddle-point search method. Then, the new saddle-points are connected to the minima by gradient descent. A numerical continuation algorithm from the computational chemistry literature is used to find saddle-points. It traces the valleys of the potential field, which are gradient extremal paths, defined as the points where the gradient is an eigenvector of the Hessian matrix. The definition of gradient bisectors is also discussed. The presentation conclude simulations in cluttered environments.
Full-text available
We consider a class of multi-robot motion planning problems where each robot is associated with multiple objectives and decoupled task specifications. The problems are formulated as an open-loop non-cooperative differential game. A distributed anytime algorithm is proposed to compute a Nash equilibrium of the game. The following properties are proven: (i) the algorithm asymptotically converges to the set of Nash equilibrium; (ii) for scalar cost functionals, the price of stability equals one; (iii) for the worst case, the computational complexity and communication cost are linear in the robot number.
Full-text available
The probabilistic roadmap approach is one of the leading motion planning techniques. Over the past eight years the technique has been studied by many different researchers. This has led to a large number of variants of the approach, each with its own merits. It is difficult to compare the different techniques because they were tested on different types of scenes, using different underlying libraries, implemented by different people on different machines. In this paper we provide a comparative study of a number of these techniques, all implemented in a single system and run on the same test scenes and on the same computer. In particular we compare collision checking techniques, basic sampling techniques, and node adding techniques. The results should help future users of the probabilistic roadmap planning approach to choose the correct techniques.
Full-text available
Although pedestrians have individual preferences, aims, and destinations, the dynamics of pedestrian crowds is surprisingly predictable. Pedestrians can move freely only at small pedestrian densities. Otherwise their motion is affected by repulsive interactions with other pedestrians, giving rise to self-organization phenomena. Examples of the resulting patterns of motion are separate lanes of uniform walking direction in crowds of oppositely moving pedestrians or oscillations of the passing direction at bottlenecks. If pedestrians leave footprints on deformable ground (for example, in green spaces such as public parks) this additionally causes attractive interactions which are mediated by modifications of their environment. In such cases, systems of pedestrian trails will evolve over time. The corresponding computer simulations are a valuable tool for developing optimized pedestrian facilities and way systems.
Conference Paper
Full-text available
This paper presents a new method for planning motions of multi-arm systems in constrained workspaces, for which state-of-the-art planners behave poorly. The method is based on the decomposition of the system into parts. Compact roadmaps are first computed for each part, and then, a super-graph is constructed by the composition of elementary roadmaps. Results presented for a three-arm system and a model of the complex DLR's Justin robot show a significant performance gain of such a two-stage roadmap construction method with respect to single-stage methods applied to the whole system.
Full-text available
Many human social phenomena, such as cooperation, the growth of settlements, traffic dynamics and pedestrian movement, appear to be accessible to mathematical descriptions that invoke self-organization. Here we develop a model of pedestrian motion to explore the evolution of trails in urban green spaces such as parks. Our aim is to address such questions as what the topological structures of these trail systems are, and whether optimal path systems can be predicted for urban planning. We use an 'active walker' model that takes into account pedestrian motion and orientation and the concomitant feedbacks with the surrounding environment. Such models have previously been applied to the study of complex structure formation in physical, chemical and biological systems. We find that our model is able to reproduce many of the observed large-scale spatial features of trail systems.
Conference Paper
Full-text available
The coordination of robot motions is one of the fundamental problems for multi-robot systems. A popular approach to avoid planning in the high-dimensional composite configuration space is the prioritized and decoupled technique. In this paper we present a method for optimizing priority schemes for such prioritized and decoupled planning technique. Our approach performs a randomized search with hill-climbing to find solutions and to minimize the overall path lengths. The technique has been implemented and tested on real robots and in extensive simulation runs. The experimental results demonstrate that our method is able to greatly reduce the number of failures and to significantly reduce the overall path length for different prioritized and decoupled path planning techniques and even for large teams of robots.
Full-text available
A new motion planning method for robots in static workspaces is presented. This method proceeds in two phases: a learning phase and a query phase. In the learning phase, a probabilistic roadmap is constructed and stored as a graph whose nodes correspond to collision-free configurations and whose edges correspond to feasible paths between these configurations. These paths are computed using a simple and fast local planner. In the query phase, any given start and goal configurations of the robot are connected to two nodes of the roadmap; the roadmap is then searched for a path joining these two nodes. The method is general and easy to implement. It can be applied to virtually any type of holonomic robot. It requires selecting certain parameters (e.g., the duration of the learning phase) whose values depend on the scene, that is the robot and its workspace. But these values turn out to be relatively easy to choose, Increased efficiency can also be achieved by tailoring some components of the method (e.g., the local planner) to the considered robots. In this paper the method is applied to planar articulated robots with many degrees of freedom. Experimental results show that path planning can be done in a fraction of a second on a contemporary workstation (≈150 MIPS), after learning for relatively short periods of time (a few dozen seconds)
Conference Paper
In this paper we address the demand for flexibility and economic efficiency in industrial autonomous guided vehicle (AGV) systems by the use of cloud computing. We propose a cloud-based architecture that moves parts of mapping, localization and path planning tasks to a cloud server. We use a cooperative longterm Simultaneous Localization and Mapping (SLAM) approach which merges environment perception of stationary sensors and mobile robots into a central Holistic Environment Model (HEM). Further, we deploy a hierarchical cooperative path planning approach using Conflict-Based Search (CBS) to find optimal sets of paths which are then provided to the mobile robots. For communication we utilize the Manufacturing Service Bus (MSB) which is a component of the manufacturing cloud platform Virtual Fort Knox (VFK). We demonstrate the feasibility of this approach in a real-life industrial scenario. Additionally, we evaluate the system's communication and the planner for various numbers of agents.
This paper presents a novel integrated approach for efficient optimization based online trajectory planning of topologically distinctive mobile robot trajectories. Online trajectory optimization deforms an initial coarse path generated by a global planner by minimizing objectives such as path length, transition time or control effort. Kinodynamic motion properties of mobile robots and clearance from obstacles impose additional equality and inequality constraints on the trajectory optimization. Local planners account for efficiency by restricting the search space to locally optimal solutions only. However, the objective function is usually non-convex as the presence of obstacles generates multiple distinctive local optima. The proposed method maintains and simultaneously optimizes a subset of admissible candidate trajectories of distinctive topologies and thus seeking the overall best candidate among the set of alternative local solutions. Time-optimal trajectories for differential-drive and carlike robots are obtained efficiently by adopting the Timed-Elastic-Band approach for the underlying trajectory optimization problem. The investigation of various example scenarios and a comparative analysis with conventional local planners confirm the advantages of integrated exploration, maintenance and optimization of topologically distinctive trajectories.
This paper deals with a holistic approach to coordinate a fleet of automated guided vehicles (AGVs) in an industrial environment. We propose an ensemble approach based on a two layer control architecture and on an automatic algorithm for the definition of the roadmap. The roadmap is built by considering the path planning algorithm implemented on the hierarchical architecture and vice versa. In this way, we want to manage the coordination of the whole system in order to increase the flexibility and the global efficiency. Furthermore, the roadmap is computed in order to maximize the redundancy, the coverage and the connectivity. The architecture is composed of two layers. The low-level represents the roadmap itself. The high-level describes the topological relationship among different areas of the environment. The path planning algorithm works on both these levels and the subsequent coordination among AGVs is obtained exploiting shared resource (i.e., centralized information) and local negotiation (i.e., decentralized coordination). The proposed approach is validated by means of simulations and comparison using real plants.
In the multi-agent pathfinding problem (MAPF) we are given a set of agents each with respective start and goal positions. The task is to find paths for all agents while avoiding collisions. Most previous work on solving this problem optimally has treated the individual agents as a single ‘joint agent’ and then applied single-agent search variants of the A* algorithm. In this paper we present the Conflict Based Search (CBS) a new optimal multi-agent pathfinding algorithm. CBS is a two-level algorithm that does not convert the problem into the single ‘joint agent’ model. At the high level, a search is performed on a Conflict Tree (CT) which is a tree based on conflicts between individual agents. Each node in the CT represents a set of constraints on the motion of the agents. At the low level, fast single-agent searches are performed to satisfy the constraints imposed by the high level CT node. In many cases this two-level formulation enables CBS to examine fewer states than A* while still maintaining optimality. We analyze CBS and show its benefits and drawbacks. Additionally we present the Meta-Agent CBS (MA-CBS) algorithm. MA-CBS is a generalization of CBS. Unlike basic CBS, MA-CBS is not restricted to single-agent searches at the low level. Instead, MA-CBS allows agents to be merged into small groups of joint agents. This mitigates some of the drawbacks of basic CBS and further improves performance. In fact, MA-CBS is a framework that can be built on top of any optimal and complete MAPF solver in order to enhance its performance. Experimental results on various problems show a speedup of up to an order of magnitude over previous approaches.
Conference Paper
Cooperative pathfinding is a problem of finding a set of non-conflicting trajectories for a number of mobile agents. Its applications include planning for teams of mobile robots, such as autonomous aircrafts, cars, or underwater vehicles. The state-of-the-art algorithms for cooperative pathfinding typically rely on some heuristic forward-search pathfinding technique, where A* is often the algorithm of choice. Here, we propose MA-RRT*, a novel algorithm for multi-agent path planning that builds upon a recently proposed asymptotically-optimal sampling-based algorithm for finding single-agent shortest path called RRT*. We experimentally evaluate the performance of the algorithm and show that the sampling-based approach offers better scalability than the classical forward-search approach in relatively large, but sparse environments, which are typical in real-world applications such as multi-aircraft collision avoidance.
Conference Paper
In this paper, the generalized motion planning algorithm (Generalized PRM: GRPM [1, 2, 3, 4]) is extended to a class of multi-agent motion planning problem in presence of process uncertainty and stochastic maps. The proposed algorithm is a hierarchical approach towards constructing a passive coordination strategy which utilizes an existing multiple traveling salesman problem (MTSP) solution methodology in conjunction with the GPRM framework to solve the multi-agent motion planning problem. The proposed algorithm is generalized to tackle multi-agent problems involving heterogeneous agents. The algorithm is used to solve multi-agent motion planning problems involving 2-dimensional (2D) and 3-dimensional(3D) agents in stochastic maps with uncertainty in the motion model. Results indicate that the algorithm successfully solves the problem under uncertainty, and generates a solution having high probability of success. It also demonstrates that the algorithm is scalable in terms of number of start and goal locations, the number of agents and their dynamics.
This paper studies a distributed path-planning problem: How can a sensor network help navigate a nontrial robot to its desired goal in a distributed manner? We consider the case where each sensor node is equipped with sophisticated sensors capable of giving a map for its sensing region. We propose a distributed sampling-based planning algorithm, where every sensor node creates a local roadmap in its locally sensed environment; these local roadmaps are “stitched” together by passing messages among nodes and forming a larger implicit roadmap without having a global representation. Based on the implicit roadmap, a feasible path is computed in a distributed manner, and the robot moves along the path by interacting with sensor nodes, each of which giving a portion of the path within the local environment of the node. Simulations show that the algorithm is able to solve the path-planning problem with low communication overhead.
Conference Paper
This paper describes a navigation system that allowed a robot to complete 26.2 miles of autonomous navigation in a real office environment. We present the methods required to achieve this level of robustness, including an efficient Voxel-based 3D mapping algorithm that explicitly models unknown space. We also provide an open-source implementation of the algorithms used, as well as simulated environments in which our results can be verified.
Self-organization, that is to say, the phenomena by which a system self-organizes its internal structure independent of external causes, is a fundamental property of open and complex systems. Such systems also exhibit phenomena of nonlinearity, instability, fractal structures and chaos — phenomena which are intimately related to the general sensation of life and urbanism at the end of the 20th century. On the other hand, self-organization is a formal theory. It is, in fact, a general umbrella for several theoretical approaches which, while they agree on general principles, differ in their treatment of such systems, in the emphasis they give to the various processes and properties and in the subject matter they refer to. In this paper I discuss some of those theories and methodologies of self-organization which were applied to the domain of cities and urbanism. The discussion proceeds under the title of seven categories of cities which are related to general theories or specific methodologies: dissipative cities, synergetic cities, chaotic cities, fractal cities, cellular automata cities, sandpile cities, and FACS and IRN cities. The discussion of each category of cities starts with a short introduction to the general principles of the approach and then elaborates on its self-organizing city.
Conference Paper
In this paper we present a method for navigating a multi-robot system through an environment while additionally maintaining a predefined set of constraints. Possible constraints are the requirement to keep up the direct line-of-sight between robots or to ensure that robots stay within a certain distance. Our approach is based on graph structures that model movements and constraints separately, in order to cover different robots and a large class of possible constraints. Additionally, the partition of movement and constraint graph allows us to use known graph algorithms like Steiner trees to solve the problem of finding a target configuration for the robots. We construct so called separated distance graphs from the Steiner tree and the underlying roadmap graph, which allow assembling valid navigation plans fast.
Stage is a C++ software library that simulates multiple mobile robots. Stage version 2, as the simulation backend for the Player/Stage system, may be the most commonly used robot simulator in research and university teaching today. Development of Stage version 3 has focused on improving scalability, usability, and portability. This paper examines Stage’s scalability. We propose a simple benchmark for multi-robot simulator performance, and present results for Stage. Run time is shown to scale approximately linearly with population size up to 100,000 robots. For example, Stage simulates 1 simple robot at around 1,000 times faster than real time, and 1,000 simple robots at around real time. These results suggest that Stage may be useful for swarm robotics researchers who would otherwise use custom simulators, with their attendant disadvantages in terms of code reuse and transparency.
The parameters in a statistical theory of multiple-lane traffic have been determined from two independent sets of data. The numerical values of the parameters calculated by four different methods of estimation are in essential agreement with one another. The data suggest the important role of adaptive human behavior in determining the characteristics of congested flow.
Conference Paper
We evaluate the use of dynamic roadmaps for online motion planning in changing environments. When changes are detected in the workspace, the validity state of affected edges and nodes of a precompiled roadmap are updated accordingly. We concentrate in this paper on analyzing the tradeoffs between maintaining dynamic roadmaps and applying an on-line bidirectional rapidly-exploring random tree (RRT) planner alone, which requires no preprocessing or maintenance. We ground the analysis in several benchmarks in virtual environments with randomly moving obstacles. Different robotics structures are used, including a 17 degrees of freedom model of NASA's Robonaut humanoid. Our results show that dynamic roadmaps can be both faster and more capable for planning difficult motions than using on-line planning alone. In particular, we investigate its scalability to 3D workspaces and higher dimensional configurations spaces, as our main interest is the application of the method to interactive domains involving humanoids.
Although the problem of determining the minimum cost path through a graph arises naturally in a number of interesting applications, there has been no underlying theory to guide the development of efficient search procedures. Moreover, there is no adequate conceptual framework within which the various ad hoc search strategies proposed to date can be compared. This paper describes how heuristic information from the problem domain can be incorporated into a formal mathematical theory of graph searching and demonstrates an optimality property of a class of search strategies.
A multi-tree extension of the transition-based RRT: Application to ordering-and-pathfinding problems in continuous cost spaces
  • Didier Devaurs
  • Simeon Thierry
  • Juan Cortes
Didier Devaurs, Thierry Simeon, and Juan Cortes. 2014. A multi-tree extension of the transition-based RRT: Application to ordering-and-pathfinding problems in continuous cost spaces. In 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2991-2996. 6942975
Adam: A Method for Stochastic Optimization
  • P Diederik
  • Jimmy Lei Kingma
  • Ba
Diederik P Kingma and Jimmy Lei Ba. 2015. Adam: A Method for Stochastic Optimization. In 3rd International Conference for Learning Representations (ICLR). arXiv:1412.6980
The walking behaviour of pedestrian social groups and its impact on crowd dynamics
  • Mehdi Moussaïd
  • Niriaska Perozo
  • Simon Garnier
  • Dirk Helbing
  • Guy Theraulaz
Mehdi Moussaïd, Niriaska Perozo, Simon Garnier, Dirk Helbing, and Guy Theraulaz. 2010. The walking behaviour of pedestrian social groups and its impact on crowd dynamics. PLoS ONE 5, 4 (apr 2010), e10047. https: // arXiv:1003.3894
Finding a needle in an exponential haystack: Discrete RRT for exploration of implicit roadmaps in multirobot motion planning
  • Kiril Solovey
  • Oren Salzman
  • Dan Halperin
Kiril Solovey, Oren Salzman, and Dan Halperin. 2016. Finding a needle in an exponential haystack: Discrete RRT for exploration of implicit roadmaps in multirobot motion planning. The International Journal of Robotics Research 35, 5 (apr 2016), 501-513.
Real-time navigation of independent agents using adaptive roadmaps
  • Avneesh Sud
  • Russell Gayle
  • Erik Andersen
  • Stephen Guy
  • Ming Lin
  • Dinesh Manocha
Avneesh Sud, Russell Gayle, Erik Andersen, Stephen Guy, Ming Lin, and Dinesh Manocha. 2008. Real-time navigation of independent agents using adaptive roadmaps. In ACM SIGGRAPH 2008 classes on -SIGGRAPH '08. ACM Press, New York, New York, USA, 1.
Towards a probabilistic roadmap for multi-robot coordination
  • Z Yan
  • A A Jouandeau
  • Cherif
Z Yan, N Jouandeau, and A A Cherif. 2012. Towards a probabilistic roadmap for multi-robot coordination. Proceedings of the 2012 International Conference on Artificial Intelligence, ICAI 2012 2 (2012), 611-617.
  • Jingjin Yu
  • Steven M Lavalle
Jingjin Yu and Steven M. LaValle. 2015. Optimal Multirobot Path Planning on Graphs: Complete Algorithms and Effective Heuristics. 32, 5 (2015), 1163-1177. arXiv:1507.03289