This paper introduces the Proactive Assistance through action-Completion Estimation (PACE) framework, designed to enhance human-robot collaboration through real-time monitoring of human progress. PACE incorporates a novel method that combines Dynamic Time Warping (DTW) with correlation analysis to track human task progression from hand movements. PACE trains a reinforcement learning policy from limited demonstrations to generate a proactive assistance policy that synchronizes robotic actions with human activities, minimizing idle time and enhancing collaboration efficiency. We validate the framework through user studies involving 12 participants, showing significant improvements in interaction fluency, reduced waiting times, and positive user feedback compared to traditional methods.

Behavior Trees (BTs) were first conceived in the computer games industry as a tool to model agent behavior, but they received interest also in the robotics community as an alternative policy design to Finite State Machines (FSMs). The advantages of BTs over FSMs had been highlighted in many works, but there is no thorough practical comparison of the two designs. Such a comparison is particularly relevant in the robotic industry, where FSMs have been the state-of-the-art policy representation for robot control for many years. In this work we shed light on this matter by comparing how BTs and FSMs behave when controlling a robot in a mobile manipulation task. The comparison is made in terms of reactivity, modularity, readability, and design. We propose metrics for each of these properties, being aware that while some are tangible and objective, others are more subjective and implementation dependent. The practical comparison is performed in a simulation environment with validation on a real robot. We find that although the robot’s behavior during task solving is independent on the policy representation, maintaining a BT rather than an FSM becomes easier as the task increases in complexity.

In this paper we provide a practical demonstration of how the modularity in a Behavior Tree (BT) decreases the effort in programming a robot task when compared to a Finite State Machine (FSM). In recent years the way to represent a task plan to control an autonomous agent has been shifting from the standard FSM towards BTs. Many works in the literature have highlighted and proven the benefits of such design compared to standard approaches, especially in terms of modularity, reactivity and human readability. However, these works have often failed in providing a tangible comparison in the implementation of those policies and the programming effort required to modify them. This is a relevant aspect in many robotic applications, where the design choice is dictated both by the robustness of the policy and by the time required to program it. In this work, we compare backward chained BTs with a fault-tolerant design of FSMs by evaluating the cost to modify them. We validate the analysis with a set of experiments in a simulation environment where a mobile manipulator solves an item fetching task.

Modern industrial applications require robots to operate in unpredictable environments, and programs to be created with a minimal effort, to accommodate frequent changes to the task. Here, we show that genetic programming can be effectively used to learn the structure of a behavior tree (BT) to solve a robotic task in an unpredictable environment. We propose to use a simple simulator for learning, and demonstrate that the learned BTs can solve the same task in a realistic simulator, converging without the need for task specific heuristics, making our method appealing for real robotic applications.

Reinforcement Learning (RL) of robotic manipulation skills, despite its impressive successes, stands to benefit from incorporating domain knowledge from control theory. One of the most important properties that is of interest is control stability. Ideally, one would like to achieve stability guarantees while staying within the framework of state-of-the-art deep RL algorithms. Such a solution does not exist in general, especially one that scales to complex manipulation tasks. We contribute towards closing this gap by introducing normalizing-flow control structure, that can be deployed in any latest deep RL algorithms. While stable exploration is not guaranteed, our method is designed to ultimately produce deterministic controllers with provable stability. In addition to demonstrating our method on challenging contact-rich manipulation tasks, we also show that it is possible to achieve considerable exploration efficiency–reduced state space coverage and actuation efforts– without losing learning efficiency.

In this paper, we introduce the problem of cross-modal visuo-tactile object recognition with robotic active exploration. With this term, we mean that the robot observes a set of objects with visual perception, and later on, it is able to recognize such objects only with tactile exploration, without having touched any object before. Using a machine learning terminology, in our application, we have a visual training set and a tactile test set, or vice versa. To tackle this problem, we propose an approach constituted by four steps: finding a visuo-tactile common representation, defining a suitable set of features, transferring the features across the domains, and classifying the objects. We show the results of our approach using a set of 15 objects, collecting 40 visual examples and five tactile examples for each object. The proposed approach achieves an accuracy of 94.7%, which is comparable with the accuracy of the monomodal case, i.e., when using visual data both as training set and test set. Moreover, it performs well compared to the human ability, which we have roughly estimated carrying out an experiment with ten participants.

In this work, we propose a framework to deal with cross-modal visuo-tactile object recognition. By cross-modal visuo-tactile object recognition, we mean that the object recognition algorithm is trained only with visual data and is able to recognize objects leveraging only tactile perception. The proposed cross-modal framework is constituted by three main elements. The first is a unified representation of visual and tactile data, which is suitable for cross-modal perception. The second is a set of features able to encode the chosen representation for classification applications. The third is a supervised learning algorithm, which takes advantage of the chosen descriptor. In order to show the results of our approach, we performed experiments with 15 objects common in domestic and industrial environments. Moreover, we compare the performance of the proposed framework with the performance of 10 humans in a simple cross-modal recognition task.

In this work, we propose a model-based and data efficient approach for reinforcement learning. The main idea of our algorithm is to combine simulated and real rollouts to efficiently find an optimal control policy. While performing rollouts on the robot, we exploit sensory data to learn a probabilistic model of the residual difference between the measured state and the state predicted by a simplified model. The simplified model can be any dynamical system, from a very accurate system to a simple, linear one. The residual difference is learned with Gaussian processes. Hence, we assume that the difference between real and simplified model is Gaussian distributed, which is less strict than assuming that the real system is Gaussian distributed. The combination of the partial model and the learned residuals is exploited to predict the real system behavior and to search for an optimal policy. Simulations and experiments show that our approach significantly reduces the number of rollouts needed to find an optimal control policy for the real system.

In this work, we propose a Frequency-based Action Descriptor (FADE) to represent human actions. In robotics, with the development of Programming by Demonstration (PbD) methods, representing and recognizing large sets of actions has become crucial to build autonomous systems that learn from humans. The FADE descriptor leverages Fast Fourier Transform (FFT) for action representation and is combined with the Manhattan distance for measuring similarities between actions. It is characterized by a low time and space complexity and is particularly suitable for classification of human actions. For clustering problems, we propose a modified version of FADE, called Uncompressed-FADE (U-FADE), which performs well in combination with Spectral Clustering algorithms at the price of a reduced compression. We compare FADE with action descriptors based on Singular Value Decomposition (SVD) and Hidden Markov Models (HMM) on the entire HDM05 motion capture database. Despite the high dimensionality of the problem, we obtained on the entire database a promising recognition rate of 78% combining FADE with a simple 1-NN classification algorithm. Furthermore, we achieved a rate of 98% on a small action set and 88% on a medium action set.

This paper proposes an experimental study of slipping avoidance algorithms based on force/tactile perception data. The claim is that contact force measurements alone or tactile data alone are not sufficient for an effective slipping avoidance strategy in real world conditions. Integrated force/tactile sensors able to provide measurements of both the contact force vector and spatially distributed tactile maps are the key enabling technology for efficient slipping avoidance control algorithms that can actually work with real world objects under no restricting assumption on the contact geometry or with unknown physical properties of the objects. The paper proposes a new slipping avoidance control scheme, which usefully exploits an integrated force/tactile sensor mounted on the parallel gripper of a Kuka youBot. The results show how the strategy successfully allows the robot to safely manipulate real-world objects, both rigid and compliant, in various friction conditions of their surface, both stable and slippery.

This paper presents an automated method for segmentation and classification of manipulation tasks. It introduces a method to build and update a dictionary of elementary actions, so as to express observed tasks as a sequence of items. Segmentation is carried out by splitting an observed manipulation task into submaneuvers. It is based on singular value decomposition of data that is gathered from the observation of humans. This observation consists of hand joint angles, the hand pose with respect to a world frame, and fingertip contact forces. The classification step introduces, from a large set of observed maneuvers, new entities called elementary actions that generalize the concept of segments, instances of elementary actions. This paper uses fingertip contact forces in the measured data. In grasping and manipulation tasks, the interaction between the hand and the object in the physical world is necessary to segment and interpret motion. A set of$\hbox{120}$maneuvers involving six tasks have been used to evaluate the methods with dependent measures including metrics of robustness, effectiveness, and repeatability. In such evaluations, the average value of the effectiveness metrics over all the maneuvers is$\hbox{0.866}$. The interuser repeatability is equal to$\hbox{0.8926}$, while the average repeatability is$\hbox{0.911}$.

The stability of discrete time kinematic sensor-based control of robots is investigated in this paper. A hierarchical inner-loop/outer-loop control architecture common for a generic robotic system is considered. The inner loop is composed of a servo-level joint controller and higher level kinematic feedback is performed in the outer loop. Stability results derived in this paper are of interest in several applications including visual servoing problems, redundancy control, and coordination/synchronization problems. The stability of the overall system is investigated taking into account input/output delays and the inner loop dynamics. A necessary and sufficient condition that the gain of the outer feedback loop has to satisfy to ensure local stability is derived. Experiments on a Kuka K-R16 manipulator have been performed in order to validate the theoretical findings on a real robotic system and show their practical relevance.

The aim of this paper is to investigate the discrete-time stability of robot motion control in the task space. The control system has been modeled as a classical inner-loop/outer-loop architecture, adopted in several industrial robotic systems. The inner-loop is composed of a servo-level joint controller, and higher level kinematic feedback is performed in the outer-loop. Heterogeneous dynamics is considered in the inner-loop, which can for instance describe redundant coordination/synchronization control systems with cooperative robots with non-identical dynamical responses. There are surprisingly few discrete-time stability results in the current state-of-the-art for this popular control architecture. The qualitative effects of the inner-loop dynamics on the overall stability of the system is investigated, and improved outer-loop feedback gain margins are derived.

The purpose of this paper is to provide a convergence analysis of classical inverse kinematics algorithms for redundant robots, whose stability is usually proved only in the continuous-time domain, thus neglecting limits of the actual implementation in the discrete time, whereas the convergence analysis carried out in this paper in the discrete-time domain provides a method to find bounds on the gain of the closed-loop inverse kinematics algorithms in relation to the sampling time. It also provides an estimation of the region of attraction (without resorting to Lyapunov arguments), i.e., upper bounds on the initial task space error. Simulations on an 11-degree-of-freedom manipulator are performed to show how the found bounds on the gain are not too restrictive.