Automatic Deployment of Distributed Teams of Robots From Temporal Logic Motion Specifications

Kloetzer, Marius; Belta, Călin

doi:10.1109/tro.2009.2035776

Cited by 156 publications

(153 citation statements)

References 25 publications

Supporting

Mentioning

152

Contrasting

Unclassified

Order By: Relevance

“…LTL is widely used in software engineering and program verification, but it has also received attention from the autonomous robotics community since the 1990s-see [22]. So far, LTL has been applied to declarative specification and synthesis of controllers from a high-level specification [23] and it is also applied for motion planning [24][25][26]. Similar to the approach described in this paper, Kreutzmann et al [27] apply LTL to specify processes in logistics declaratively.…”

Section: Spatial Logic Of Manipulation Tasksmentioning

confidence: 99%

Leveraging Qualitative Reasoning to Learning Manipulation Tasks

Wolter

Kirsch

2015

Robotics

View full text Add to dashboard Cite

Learning and planning are powerful AI methods that exhibit complementary strengths. While planning allows goal-directed actions to be computed when a reliable forward model is known, learning allows such models to be obtained autonomously. In this paper we describe how both methods can be combined using an expressive qualitative knowledge representation. We argue that the crucial step in this integration is to employ a representation based on a well-defined semantics. This article proposes the qualitative spatial logic QSL, a representation that combines qualitative abstraction with linear temporal logic, allowing us to represent relevant information about the learning task, possible actions, and their consequences. Doing so, we empower reasoning processes to enhance learning performance beyond the positive effects of learning in abstract state spaces. Proof-of-concept experiments in two simulation environments show that this approach can help to improve learning-based robotics by quicker convergence and leads to more reliable action planning.Keywords: qualitative spatial reasoning; robot learning; AI robotics MotivationRobotic applications have evolved considerably, yet they still lack the efficiency, flexibility and adaptability that is needed to improve our everyday life as versatile service robots. Planning and learning Robotics 2015, 4 254 are the two major paradigms employed to make a robot act intelligently. Planning is a form of symbolic reasoning applied on various levels of abstraction. Planning is effective whenever reliable forward models of the robot and its environment are available. Learning is a powerful method to generate such models from data. However, it requires carefully handcrafted state space representations to converge to high-quality models in reasonable time. Neither planning nor learning on their own suffice to meet the demands of versatile service robots. By integrating both paradigms one can benefit from their respective strengths. So far, a true integration of learning and planning has not been achieved that allows a robot to acquire new skills efficiently in a truly autonomous way.Our work is motivated by this basic question of how to integrate learning and planning, leveraging their individual strengths to produce reliable and versatile robot behavior. We propose a qualitative reasoning framework as the glue between those two paradigms. Qualitative reasoning lifts information, e.g., obtained by perception, to a knowledge level, enabling us to integrate it with common sense knowledge and to use available reasoning techniques [1][2][3]. The qualitative representation serves two purposes: (1) to make a planning or action selection process with learned models more efficient; and (2) to control selection of learning samples in order to make a learning process more reliable and to obtain high-quality models with little manual design effort. We present two experimental studies in simulation for a specific robot task: to throw an object to a specific destination. The results ind...

show abstract

Section: Spatial Logic Of Manipulation Tasksmentioning

confidence: 99%

Leveraging Qualitative Reasoning to Learning Manipulation Tasks

Wolter

Kirsch

2015

Robotics

View full text Add to dashboard Cite

show abstract

“…Since each PTZ camera has an embedded controller, it is possible to design a decentralized control protocol. This can be done, for instance, by incorporating some communication constraints into the centralized design as in [23] so that the resulting controllers can be implemented on individual PTZs. However, a more efficient approach would be not only to decentralize the implementation of the control protocol but also to distribute the computations in the synthesis which we discuss next.…”

Section: A Centralized Control Protocol Synthesismentioning

confidence: 99%

Distributed Synthesis of Control Protocols for Smart Camera Networks

Özay

Topcu

Murray

et al. 2011

2011 IEEE/ACM Second International Conference on Cyber-Physical Systems

View full text Add to dashboard Cite

Abstract-We considered the problem of designing control protocols for pan-tilt-zoom (PTZ) cameras within a smart camera network where the goal is to guarantee certain temporal logic specifications related to a given surveillance task. We first present a centralized control architecture for assigning PTZ cameras to targets so that the specification is met for any admissible behavior of the targets. Then, in order to alleviate the computational complexity associated with LTL synthesis and to enable implementation of local control protocols on individual PTZ cameras, we propose a distributed synthesis methodology. The main idea is to decompose the global specification into local specifications for each PTZ camera. These decompositions allow the protocols for each camera to be separately synthesized and locally implemented while guaranteeing the global specifications to hold. A thorough design example is presented to illustrate the steps of the proposed procedure.

show abstract

“…[15,26,24] use Linear Temporal Logic to formally describe uncertain multi-agent robotics by a finite state partitioning of the 2D environment. We adopt a discrete representation more suitable to high dimensional spaces; our manipulation task uses a 7-DOF robot and 40 movable objects making complete discretization computationally infeasible.…”

Section: Related Work Literature On Grammars From the Linguistic mentioning

confidence: 99%

The Motion Grammar: Linguistic Perception, Planning, and Control

Dantam¹,

Stilman²

2011

Robotics: Science and Systems VII

View full text Add to dashboard Cite

Abstract-We present and analyze the Motion Grammar: a novel unified representation for task decomposition, perception, planning, and control that provides both fast online control of robots in uncertain environments and the ability to guarantee completeness and correctness. The grammar represents a policy for the task which is parsed in real-time based on perceptual input. Branches of the syntax tree form the levels of a hierarchical decomposition, and the individual robot sensor readings are given by tokens. We implement this approach in the interactive game of Yamakuzushi on a physical robot resulting in a system that repeatably competes with a human opponent in sustained gameplay for the roughly six minute duration of each match. I. INTRODUCTIONAs robots come into increasing contact with humans, it is absolutely vital to prove that these potentially dangerous machines are safe and reliable. Furthermore, applying robots to increasingly complicated and varied tasks requires a tractable way to generate desired behaviors. Typical reactive strategies focus on efficiency rather than proving how the system will respond during complicated tasks with uncertain outcomes [29,2]. Existing deliberative planners do provide guarantees but often simplify the control system or have prohibitive computational cost [31]. By representing perception, planning, and control of robotic systems using Context-Free Grammars (CFG), the Motion Grammar (MG) [9] enables robots to handle uncertainty in the outcomes of control actions through online parsing. Furthermore, MG makes it possible to prove that the system will correctly respond to all possible situations.Imagine a robot searching for earthquake survivors. This robot must carefully remove pieces of rubble while ensuring that the larger structure does not collapse. It must also coordinate its efforts with other robots and human rescuers. We consider a simplified version of this scenario in the two-player game of Yamakuzushi. While the game is adversarial, the robot and human collaborate to safely disassemble a pile of Shogi pieces. The game contains many of the challenging elements of physical human-robot interaction. Both the robot and human make careful contact with pieces without causing them to fall. The robot slides pieces without losing contact. It also observes human actions and handles dangerous conditions. The key to all these interactions is that the robot must handle uncertainty. Our Motion Grammar guarantees that the robot responds to the complete range of uncertain outcomes online.The Motion Grammar is a linguistic approach to robot control with significant benefits over existing techniques. The fast algorithms for Context-Free parsing enable the robot to quickly react online without lengthy deliberative planning. CFGs represent a balance between power and provability. This allows the system designer to tackle a broader class of problems and still prove desired response with model-checking.

show abstract

Automatic Deployment of Distributed Teams of Robots From Temporal Logic Motion Specifications

Cited by 156 publications

References 25 publications

Leveraging Qualitative Reasoning to Learning Manipulation Tasks

Leveraging Qualitative Reasoning to Learning Manipulation Tasks

Distributed Synthesis of Control Protocols for Smart Camera Networks

The Motion Grammar: Linguistic Perception, Planning, and Control

Contact Info

Product

Resources

About