Browsing by Author "He, Keliang"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Automated Abstraction of Manipulation Domains for Cost-Based Reactive Synthesis
    (IEEE, 2019) He, Keliang; Lahijanian, Morteza; Kavraki, Lydia E.; Vardi, Moshe Y.
    When robotic manipulators perform high-level tasks in the presence of another agent, e.g., a human, they must have a strategy that considers possible interferences in order to guarantee task completion and efficient resource usage. One approach to generate such strategies is called reactive synthesis. Reactive synthesis requires an abstraction, which is a discrete structure that captures the domain in which the robot and other agents operate. Existing works discuss the construction of abstractions for mobile robots through space decomposition; however, they cannot be applied to manipulation domains due to the curse of dimensionality caused by the manipulator and the objects. In this work, we present the first algorithm for automatic abstraction construction for reactive synthesis of manipulation tasks. We focus on tasks that involve picking and placing objects with possible extensions to other types of actions. The abstraction also provides an upper bound on path-based costs for robot actions. We combine this abstraction algorithm with our reactive synthesis planner to construct correct-by-construction plans. We demonstrate the power of the framework on a UR5 robot, completing complex tasks in face of interferences by a human.
  • Thumbnail Image
    Item
    Reactive Synthesis for Finite-Horizon Robot Tasks
    (2019-02-15) He, Keliang; Kavraki, Lydia E.; Vardi, Moshe Y.
    Traditionally, robots are limited to controlled environments such as caged areas in factories. When robots move to uncontrolled environments where humans are present, the robot must consider the possible behaviors of the humans. To ensure the robot completes its task, a class of approaches called reactive synthesis has been employed to construct reactive strategies for robots. The reactive strategy chooses robot actions based on the observed human behaviors, and guarantees task completion. This thesis expands the capabilities of reactive synthesis to finite-horizon tasks with resource cost constraints to manipulation domains, and to more complex problems. The algorithms in this thesis are demonstrated on a UR5 robot performing pick-and-place tasks. Existing works in robotic reactive synthesis focused on infinite-horizon tasks (e.g. surveillance) because they rely on existing reactive synthesis tools from the program synthesis community. However, many robotic tasks, such as assembly and delivery, are finite-horizon. This thesis presents a reactive synthesis framework for finite-horizon tasks. By focusing on finite-horizon domains, we not only guarantee task completion, but also minimize the robot’s resource consumption during execution. Reactive synthesis can only be directly performed on finite structures, but robotic domains are continuous and infinite. Thus an abstraction that discretizes the domain is required. Existing works in reactive synthesis have focused on navigation problems, where abstraction is achieved using workspace decomposition. This thesis presents an appropriate abstraction for manipulation domains. This abstraction allows reactive synthesis to be applied to manipulation domains. This thesis also presents an algorithm to make reactive synthesis more scalable. Previous synthesis algorithms in robotics used either explicit state or formulaic representations. Such representations fail to utilize the structure of robotic problems. This thesis presents a synthesis algorithm that represents the domains symbolically using binary decision diagrams, which are more compact and take advantage of the structure in robotic problems. Order-of-magnitude speed-ups over existing approaches are observed on robotics-related benchmarks. This speed-up allows reactive synthesis to be performed on much larger domains.
  • Thumbnail Image
    Item
    Robot Manipulation Planning Under Linear Temporal Logic Specifications
    (2015-12-07) He, Keliang; Kavraki, Lydia E; Vardi, Moshe Y; Chaudhuri, Swarat
    Automated planning for manipulation tasks is highly desirable, for it enables robot manipulators to be used by none robotics experts. This thesis presents one approach to solving manipulation planning for tasks expressed in linear temporal logic (ltl). This approach is based on the synergistic framework, which provides probabilistic completeness guarantees. Even though the synergistic framework has shown to work well for planning for ltl tasks in the navigation domain, it lacked an abstraction that can capture the high dimensionality of manipulation. This thesis enables manipulation planning using the synergistic framework by introducing a manipulation abstraction and modifying the interaction between task and motion planning in the framework. The modified framework is shown to be effectively in case studies in both simulation and physical systems. The case studies also show that the synergistic framework plans for manipulation problems more effective using the manipulation abstraction in comparison with a naive abstraction.