Steps towards model-free execution monitoring on mobile robots (original) (raw)

Model-free execution monitoring in behavior-based mobile robotics

2003

In this paper we present a model-free execution monitor for behavior-based mobile robots. By modelfree we mean that the monitoring is based only on the actual execution, without involving any predictive models of the controlled system. Model-free monitors are especially suitable for systems where it is hard to obtain adequate models. In our approach we analyze the activation levels of the different behaviors using a pattern recognition technique. Our model-free execution monitor, which is realized by radial basis function networks, is shown to give a high performance in several realistic simulations.

Model-Free Execution Monitoring in Behavior-Based Robotics

IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 2000

In the near future, autonomous mobile robots are expected to help humans by performing service tasks in many different areas, including personal assistance, transportation, cleaning, mining, or agriculture. In order to manage these tasks in a changing and partially unpredictable environment, without the aid of humans, the robot must have the ability to plan its actions and to execute them robustly and in a safe way. The robot must also have the ability to detect when the execution does not proceed as planned, and to correctly identify the causes of the failure. An execution monitoring system allows the robot to detect and classify these failures.

Model-free execution monitoring by learning from simulation

2005 IEEE International Symposium on Computational Intelligence in Robotics and Automation, Proceedings, 2005

Autonomous robots need the ability to plan their actions and to execute them robustly and in a safe way in face of a changing and partially unpredictable environment. This is especially important if we want to design autonomous robots that can safely co-habitate with humans. In order to manage this, these robots need the ability to detect when the execution does not proceed as planned, and to correctly identify the causes of the failure. An execution monitoring system is a system that allows the robot to detect and classify these failures. In this work we show that pattern recognition techniques can be applied to realize execution monitoring by classifying observed behavioral patterns into normal or faulty behaviors. The approach has been successfully tested on a real robot navigating in an office environment. Interesting, these tests show that we can train an execution monitor in simulation, and then use it in a real robot.

Semantic Knowledge-Based Execution Monitoring for Mobile Robots

Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007

We describe a novel intelligent execution monitoring approach for mobile robots acting in indoor environments such as offices and houses. Traditionally, monitoring execution in mobile robotics amounted to looking for discrepancies between the model-based predicted state of executing an action and the real world state as computed from sensing data. We propose to employ semantic knowledge as a source of information to monitor execution. The key idea is to compute implicit expectations, from semantic domain information, that can be observed at run time by the robot to make sure actions are executed correctly. We present the semantic knowledge representation formalism, and how semantic knowledge is used in monitoring. We also describe experiments run in an indoor environment using a real mobile robot.

Plan execution monitoring and control architecture for mobile robots

1995

This paper deals with the architecture and control structure of mobile robots. We decompose robot functions into modules organized according to their predefined interactions: sensor modules that accomplish various processings on data from physical sensors, effector modules that issue commands to effectors, servo-processes that establish links between perception and action to achieve closed-loop behaviors, and functional units that provide specific functionalities. These modules, and hence the robot system itself, are controlled by a control system that also enables the robot to execute missions (plans) expressed in a command language. We introduce and discuss a generic control system structure, composed of a Supervisor that interprets the plan and oversees its execution, an Executive for operating and managing robot modules and resources, a Surveillance Manager for detecting and reacting to asynchronous events, and an Error Recovery module for local plan mending and correction. Several experimental examples are given.

Execution Knowledge for Execution Monitoring: what, why, where and what for?

Despite the progress made in planning and robotics, autonomous plan execution on a robot remains challenging. One of the problems is that (classical) planners use abstract models which are disconnected from the sensor and actuation information available during execution. This connection is typically created in a non-systematic way by some system-specific execution software. In this paper we propose to explicitly represent Execution Knowledge that encodes the connection between planning models and the actual actions and observations for a given physical system. We present an execution monitoring framework in which Execution Knowledge captures the expectations about physical plan execution. A violation of these expectations indicates an execution failure.

Dependable execution control for autonomous robots

2004

This paper presents a new approach to integrate real-time execution control on autonomous systems and how such an approach integrates in their software architecture. The use of decisional autonomy is becoming more widely accepted as a solution to the increasing need to deploy complex systems (robots, satellites, etc) able to perform non trivial tasks in various environments. We present an overview of the organization of such systems. Then we explain why the increasing complexity of functional components as well as the presence of autonomy components become an obstacle to system safety and dependability. To address this issue, we propose the integration of an execution control component in the software architecture. This component is synthesized from a model of the acceptable and dangerous state using model-checking techniques. The execution controller has a generic representation of system behavior and, according to some specified system constraints, acts as a "safety bag" allowing acceptable states and avoiding forbidden ones. The controller uses an OBDD 1 like data structure which offers a bounded execution time, and which can be formally validated offline to check temporal properties. Real experimentations have been made on our autonomous mobile robots, and have confirmed it can catch in real-time design errors from the decisional components which would have lead to disastrous consequences.

Rule-based Dynamic Safety Monitoring for Mobile Robots

2016

Safety is a key challenge in robotics, in particular for mobile robots operating in an open and unpredictable environment. To address the safety challenge, various software-based approaches have been proposed, but none of them provide a clearly specified and isolated safety layer. In this paper, we propose that safety-critical concerns regarding the robot software be explicitly declared separately from the main program, in terms of externally observable properties of the software. Concretely, we use a Domain-Specific Language (DSL) to declaratively specify a set of safety-related rules that the software must obey, as well as corresponding corrective actions that trigger when rules are violated. Our DSL, integrated with ROS, is shown to be capable of specifying safety-related constraints, and is experimentally demonstrated to enforce safety behaviour in existing robot software. We believe our approach could be extended to other fields to similarly simplify safety certification.

Active Execution Monitoring Using Planning and Semantic Knowledge

2007

To cope with the dynamics and uncertainty inherent in real world environments, autonomous mobile robots need to per- form execution monitoring for verifying that their plans are executed as expected. Domain semantic knowledge has lately been proposed as a source of information to derive and mon- itor implicit expectations of executing actions. For instance, when the robot moves into an

Monitoring the execution of sensory robot programs

Fifth International Conference on Advanced Robotics 'Robots in Unstructured Environments, 1991

A system is presented, which monitors the execution of an assigned robot program, in order to detect execution errors. The assigned program is supposed to include sensor instructions, which allow to adapt the execution to variable environment conditions. In the presented system, the task information is represented in terms of a relationship between environment conditions and workcell evlution. The selection of the monitoring sensor detctions is based on the accuracy in checking the state variables and in the ambiguity in matching the sensor measures to the variables to be measured.