Stefano Caselli

Robust trajectory learning and approximation for robot programming by demonstration

Abstract Trajectory learning is a fundamental component in a robot Programming by Demonstration (PbD) system, where often the very purpose of the demonstration is to teach complex manipulation patterns. However, human demonstrations are inevitably noisy and inconsistent. This paper highlights the trajectory learning component of a PbD system for manipulation tasks encompassing the ability to cluster, select, and approximate human demonstrated trajectories. The proposed technique provides some advantages with respect to alternative approaches and is suitable for learning from both individual and multiple user demonstrations.

Leveraging on a virtual environment for robot programming by demonstration

Abstract The Programming by Demonstration paradigm promises to reduce the complexity of robot programming. Its aim is to let robot systems learn new behaviors from a human operator demonstration. In this paper, we argue that while providing demonstrations in the real environment enables teaching of general tasks, for tasks whose essential features are known a priori demonstrating in a virtual environment may improve efficiency and reduce trainer’s fatigue. We next describe a prototype system supporting Programming by Demonstration in a virtual environment and we report results obtained exploiting simple virtual tactile fixtures in pick-and-place tasks.

Grasp recognition in virtual reality for robot pregrasp planning by demonstration

This paper describes a virtual reality based programming by demonstration system for grasp recognition in manipulation tasks and robot pregrasp planning. The system classifies the human hand postures taking advantage of virtual grasping and information about the contact points and normals computed in the virtual reality environment. A pregrasp planning algorithm mimicking the human hand motion is also proposed. Reconstruction of human hand trajectories, approaching the objects in the environment, is based on NURBS curves and a data smoothing algorithm. Some experiments involving grasp classification and pregrasp planning, while avoiding obstacles in the workspace, show the viability and effectiveness of the approach

Rate modulation of soft real-time tasks in autonomous robot control systems

Due to the high number of sensors managed and need to perform complex reasoning activities, real-time control systems of autonomous robots exhibit a high potential for overload, i.e., real-time tasks missing their deadlines. In these systems overload should be regarded as a likely occurrence and hence managed accordingly. In this paper we illustrate a novel scheduling technique for adaptation of soft real-time load to available computational capacity in the context of autonomous robot control architectures. The technique is based on rate modulation of a set of periodic tasks in a range of admissible rates. The technique is shown to be easily computable and several variations in implementation are reviewed within the paper.

Parallel State Space Exploration for GSPN Models

Generalized Stochastic Petri Nets (GSPN) have gained a wide acceptance as a modeling tool for the performance analysis of concurrent systems. However, the applicability of this methodology is severely limited by the potential state space explosion phenomenon. In this paper we describe massively parallel approaches to the most computing-intensive part of the solution of GSPN models: the state space construction. The effectiveness of these parallel approaches stays, for every GSPN, in their ability to deal with very large reachability spaces in reasonable time. Both the SIMD and the MIMD programming models are considered, and examples are given using recent massively parallel processing architectures (CM-5, T3D).

Efficient exploration and recognition of convex objects based on haptic perception

The paper describes an efficient technique for recognition of convex objects from tactile sensing. The technique is based on the development of internal and external volumetric approximations of the unknown object, and exploits an effective feature selection strategy along with early pruning of incompatible objects to improve recognition performance. Exploration strategies conceived for a single-finger probing device rely on the developed volumetric approximations to guide sensing along directions where uncertainty about the explored object is larger, and can take into account exploration costs.

Trajectory clustering and stochastic approximation for robot programming by demonstration

This paper describes the trajectory learning component of a programming by demonstration (PbD) system for manipulation tasks. In case of multiple user demonstrations, the proposed approach clusters a set of hand trajectories and recovers smooth robot trajectories overcoming sensor noise and human motion inconsistency problems. More specifically, we integrate a geometric approach for trajectory clustering with a stochastic procedure for trajectory evaluation based on hidden Markov models. Furthermore, we propose a method for human hand trajectory reconstruction with NURBS curves by means of a best-fit data smoothing algorithm. Some experiments show the viability and effectiveness of the approach.

A software framework based on real-time CORBA for telerobotic systems

The technological developments in distributed systems have led to new telerobotic applications, such as virtual laboratories and remote maintenance of complex equipment. These applications must satisfy both the general requirements of distributed computing, e.g. location transparency and interoperability, and the domain-specific requirements of reconfigurability, guaranteed performance, real-time operation, and cooperation among robots and sensory systems. In this paper, we describe a software framework for distributed telerobotic systems exploiting advanced CORBA features, including Asynchronous Method Invocation and real-time priorities. The framework allows development of portable multithreaded client-server applications supporting concurrent and preemptable actions in the target robot system, and has been evaluated in a laboratory setup including a robot manipulator and two cameras accessible by multiple clients.

A Technique for Adaptive Scheduling of Soft Real-Time Tasks

A number of multimedia and process control applications can take advantage from the ability to adapt soft real-time load to available computational capacity. This capability is required, for example, to react to changed operating conditions as well as to ensure graceful degradation of an application under transient overloads. In this paper, we illustrate a novel adaptive scheduling technique based on rate modulation of a set of periodic tasks in a range of admissible rates. By casting constraints on rate ranges in a linear programming formulation, several adaptation policies can be considered, along with additional constraints reflecting various application requirements. The paper investigates the effectiveness of rate modulation strategies both on simulated task sets and on real experiments.

Interactive teaching of task-oriented robot grasps

This paper focuses on the problem of grasp stability and grasp quality analysis. An elegant way to evaluate the stability of a grasp is to model its wrench space. However, classical grasp quality measures suffer from several disadvantages, the main drawback being that they are not task related. Indeed, constructive approaches for approximating the wrench space including also task information have been rarely considered. This work presents an effective method for task-oriented grasp quality evaluation based on a novel grasp quality measure. We address the general case of multifingered grasps with point contacts with friction. The proposed approach is based on the concept of programming by demonstration and interactive teaching, wherein an expert user provides in a teaching phase a set of exemplar grasps appropriate for the task. Following this phase, a representation of task-related grasps is built. During task planning and execution, a grasp could be either submitted interactively for evaluation by a non-expert user or synthesized by an automatic planning system. Grasp quality is then assessed based on the proposed measure, which takes into account grasp stability along with its suitability for the task. To enable real-time evaluation of grasps, a fast algorithm for computing an approximation of the quality measure is also proposed. Finally, a local grasp optimization technique is described which can amend uncertainties arising in supplied grasps by non-expert users or assist in planning more valuable grasps in the neighborhood of candidate ones. The paper reports experiments performed in virtual reality with both an anthropomorphic virtual hand and a three-fingered robot hand. These experiments suggest the effectiveness and task relevance of the proposed grasp quality measure.