Trajectory representation by nonlinear scaling of dynamic movement primitives

Purpose This paper aims to solve the problem of the inability to apply learning methods for robot motion skills based on dynamic movement primitives (DMPs) in tasks with explicit environmental constraints, while ensuring the reliability of the robot system. Design/methodology/approach The authors propose a novel DMP that takes into account environmental constraints to enhance the generality of the robot motion skill learning method. First, based on the real-time state of the robot and environmental constraints, the task space is divided into different regions and different control strategies are used in each region. Second, to ensure the effectiveness of the generalized skills (trajectories), the control barrier function is extended to DMP to enforce constraint conditions. Finally, a skill modeling and learning algorithm flow is proposed that takes into account environmental constraints within DMPs. Findings By designing numerical simulation and prototype demonstration experiments to study skill learning and generalization under constrained environments. The experimental results demonstrate that the proposed method is capable of generating motion skills that satisfy environmental constraints. It ensures that robots remain in a safe position throughout the execution of generation skills, thereby avoiding any adverse impact on the surrounding environment. Originality/value This paper explores further applications of generalized motion skill learning methods on robots, enhancing the efficiency of robot operations in constrained environments, particularly in non-point-constrained environments. The improved methods are applicable to different types of robots.

This paper proposes a method for recognizing compound trajectories. Given a library of pre-learned movements described using dynamic movement primitives, our approach is able to break down an observed trajectory into its individual components, each of which is a segment of one of the movements in the library. We build on previous work that uses critical points for movement recognition and prediction, and extend it to handle trajectories comprising multiple segments from possibly different movements. Our approach assumes that each segment in the observed trajectory is either the initial segment of a new primitive or the continuation of the previous segment. Then, given a partial trajectory, our method is able to predict the most likely next target—i.e., the end-point of the movement currently being executed, if the latter is executed to the end. By using an effective search tree, our approach is able to run at execution time and provide an efficient way for action recognition and prediction, which has applications in human-robot interaction scenarios. We validate our approach both in simulation and in a human-robot interaction scenario involving the Baxter robot.

Dynamic movement primitives (DMPs) have proven to be an effective movement representation for motor skill learning. In this paper, we propose a new approach for training deep neural networks to synthesize dynamic movement primitives. The distinguishing property of our approach is that it can utilize a novel loss function that measures the physical distance between movement trajectories as opposed to measuring the distance between the parameters of DMPs that have no physical meaning. This was made possible by deriving differential equations that can be applied to compute the gradients of the proposed loss function, thus enabling an effective application of backpropagation to optimize the parameters of the underlying deep neural network. While the developed approach is applicable to any neural network architecture, it was evaluated on two different architectures based on encoder–decoder networks and convolutional neural networks. Our results show that the minimization of the proposed loss function leads to better results than when more conventional loss functions are used.

In this paper, composite dynamic movement primitives (DMPs) based on radial basis function neural networks (RBFNNs) are investigated for robots’ skill learning from human demonstrations. The composite DMPs could encode the position and orientation manipulation skills simultaneously for human-to-robot skills transfer. As the robot manipulator is expected to perform tasks in unstructured and uncertain environments, it requires the manipulator to own the adaptive ability to adjust its behaviours to new situations and environments. Since the DMPs can adapt to uncertainties and perturbation, and spatial and temporal scaling, it has been successfully employed for various tasks, such as trajectory planning and obstacle avoidance. However, the existing skill model mainly focuses on position or orientation modelling separately; it is a common constraint in terms of position and orientation simultaneously in practice. Besides, the generalisation of the skill learning model based on DMPs is still hard to deal with dynamic tasks, e.g., reaching a moving target and obstacle avoidance. In this paper, we proposed a composite DMPs-based framework representing position and orientation simultaneously for robot skill acquisition and the neural networks technique is used to train the skill model. The effectiveness of the proposed approach is validated by simulation and experiments.

Spontaneous task composition in urban computing environments based on social, spatial, and temporal aspects

A study was conducted in the semi‐arid Mbeere district, Eastern Kenya, to determine spatial and temporal variability of crop yields and climatic factors and their impacts on sustainability of semi‐arid agriculture. Spatial aspects were assessed by conducting a survey of on‐farm crop yields and using a computer model (QUEFTS) to predict maize yields from soil chemical indices. Temporal aspects were studied using time‐series data (rainfall, temperature and crop yields). The study did not find a significant correlation between farmers' actual yields and QUEFTS predictions, and soil nutrients were thus not the only factors influencing maize yields in the study area. Other yield‐reducing factors (climate and management) not accounted for in the QUEFTS model played a role. Grain yields of staple food crops were highly variable, fluctuating in time and space, and suboptimal. Particular aspects of rainfall (e.g. the short rains, for cowpeas and beans) were more important than mean annual rainfall in determining crop yields, and factors other than rainfall and soil fertility apparently played a role. The observed low grain yields, large yield gaps and high rainfall variability challenge the sustainability of these farming systems. Copyright © 2008 John Wiley & Sons, Ltd

Network-on-Chip (NoC) has emerged as an alternative interconnecting paradigm in the state-of-the-art multi-core architectures. Its capability of parallel data transfer over various paths increases the possibility of resource contention and causes network congestion. How to avoid contention, as well as to optimize performance and energy consumption becomes a great challenge for system designers. In this paper, we consider spatial and temporal aspects of communication to avoid contention. In spatial aspect, we propose to utilize overlapped communication paths to estimate the degree of contention, such that they can be minimized in mapping stage. In temporal aspect, we sequentialize data transfer with potential contention by introducing additional latency. To optimize the design concerns with the corresponding constraint model, we further provide an efficient algorithm to search for better solutions for the problem, which integrates a local search process into a genetic algorithm and applies various heuristics in initialization and evolution process. Experimentations from random and real-case benchmarks demonstrate the efficiency of our method in multi-objective optimization and the effectiveness of our techniques in avoiding network contention, while keeping performance and energy consumption optimized.

This paper presents the analysis of South Sumatera hotspots distribution pattern from 2005–2015 based on spatial and temporal aspects. The hotspot distribution data used is FIRMS MODIS Fires data. The spatial aspect is represented by the attribute of latitude and longitude as the location of the occurrence of hotspots and the temporal is represented by the date attribute. DBSCAN clustering is used to analyze spatial aspect. The results show that Tulung Selapan subdistrict is on high fire-prone region. Air Sugihan, Cengal, Bayung Lencir is on medium and the others is low. For temporal aspect, the result shows most hotspots occurred on August, September and October.

As social primates, one of the most important cognitive tasks we conduct, dozens of times a day, is to look at a face and extract the person's identity. During the last decade, the neural basis of face processing has been extensively investigated in humans with event-related potential (ERP) and functional MRI (fMRI). These two methods provide complementary information about the temporal and spatial aspects of the neural response, with ERPs allowing high temporal resolution of milliseconds but low spatial resolution of the neural generator and fMRI displaying a slow hemodynamic response but better spatial localization of the activated regions. Despite the extensive fMRI and ERP research of faces, only a few studies have assessed the relationship between the two methods and no study to date have collected simultaneous ERP and fMRI responses to face stimuli. In the current paper we will try to assess the spatial and temporal aspects of the neural response to faces by simultaneously collecting functional MRI and event-related potentials (ERP) to face stimuli. Our goals are twofold: 1) ERP and fMRI show a robust selective response to faces. In particular, two well-established face-specific phenomena, the RH superiority and the inversion effect are robustly found with both ERP and fMRI. Despite the extensive research of these effects with ERP and fMRI, it is still unknown to what extent their spatial (fMRI) and temporal (ERP) aspects are associated. In Study 1 we will employ an individual differences approach, to assess the relationship between these ERP and fMRI face-specific responses. 2) Face processing involves several stages starting from structural encoding of the face image through identity processing to storage for later retrieval. This representation undergoes several manipulations that take place at different time points and in different brain regions before the final percept is generated. By simultaneously recording ERP and fMRI we hope to gain a more comprehensive understanding of the timecourse that different brain areas participate in the generation of the face representation.

Segmenting complex movements into a sequence of primitives remains a difficult problem with many applications in the robotics and vision communities. In this work, we show how the movement segmentation problem can be reduced to a sequential movement recognition problem. To this end, we reformulate the original Dynamic Movement Primitive (DMP) formulation as a linear dynamical system with control inputs. Based on this new formulation, we develop an Expectation-Maximization algorithm to estimate the duration and goal position of a partially observed trajectory. With the help of this algorithm and the assumption that a library of movement primitives is present, we present a movement segmentation framework. We illustrate the usefulness of the new DMP formulation on the two applications of online movement recognition and movement segmentation.

This paper addresses a significant challenge in achieving collaborative tasks; how can a robot or multiple robots, endowed with a library of pre-learned primitive movements, generate multiple simultaneous coordinated robotic movements, adapting and optimizing those in the library, to complete one collaborative task? This work can thus be seen as a follow-up to the work with a motion presented as dynamic movement primitive (DMP) that now considers collaborative tasks and the existence of multiple robots/manipulators. Specifically, we start with a simple task using one DMP and extend it to accommodate the coordinated execution of multiple DMPs in robots with multiple manipulators or-alternatively-multiple robots with a single manipulator. We investigate mechanisms to jointly optimize multiple DMPs to perform one task in a coordinated fashion. The joint trajectory is built from initial DMPs learned for a single manipulator, and its optimization must comply with task-specific constraints. We illustrate the application of our approach both in a simulated environment and in a simulated and real Baxter robot.

Human-powered lower exoskeletons have gained considerable interests from both academia and industry over the past few decades, and thus have seen increasing applications in areas of human locomotion assistance and strength augmentation. One of the most important aspects in those applications is to achieve robust control of lower exoskeletons, which, in the first place, requires the proactive modeling of human movement trajectories through physical Human-Robot Interaction (pHRI). As a powerful representation tool for motion trajectories, Dynamic Movement Primitive (DMP) has been used extensively to model human movement trajectories. However, canonical DMPs only offers a general offline representation of human movement trajectory and neglects the real-time interaction term, therefore it cannot be directly applied to lower exoskeletons which need to model human motion trajectories online since different pilots have different trajectories and even one pilot might change his/her intended trajectory during walking. This paper presents a novel Coupled Cooperative Primitives (CCPs) scheme, which models the motion trajectories online. Besides maintaining canonical motion primitives, we also model the interaction term between the pilot and exoskeletons through impedance models and apply a reinforcement learning method based on Policy Improvement and Path Integrals (PI <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) to learn the parameters online. Experimental results on both a single Degree-Of-Freedom (DOF) platform and a HUman-powered Augmentation Lower EXoskeleton (HUALEX) system demonstrate the advantages of our proposed CCP scheme.

Hemiparesis has become prevalent in recent years, and robot-assisted rehabilitation provides great therapeutic potential. Robotic mirror therapy is a promising approach for hemiplegic patients by transmitting the motion of the healthy limb (HL) to the impaired limb (IL), where the wearable robot assists the IL to mimic the HL's action to stimulate the active participation of the injured muscles. However, complete replication of the HL's movement trajectory without considering the IL's muscle strength is not satisfactory for rehabilitation. In this article, a motion generation scheme is proposed for robotic mirror therapy. The robot's movement trajectory is modeled with dynamic movement primitive (DMP), and the physical human–robot interaction is formulated as an impedance model and coupled with the DMP model. The subject's muscle strength is evaluated with skin surface electromyography signals and transferred to the robot's joint stiffness. In order to adapt the human–robot coupled DMP model parameters to varying subject performance, a reinforcement learning algorithm is designed to optimize the robot's movement trajectory and stiffness profile, where the training safety and rehabilitation improvement are both guaranteed. The proposed method is validated using a lower extremity rehabilitation robot, and experimental results demonstrate its feasibility and superiority.

Movement generation, and especially generalization to unseen situations, plays an important role in robotics. Different types of movement generation methods exist such as spline based methods, dynamical system based methods, and methods based on Gaussian mixture models (GMMs). Using a large, new dataset on human manipulations, in this paper we provide a highly detailed comparison of five fundamentally different and widely used movement encoding and generation frameworks: dynamic movement primitives (DMPs), time based Gaussian mixture regression (tbGMR), stable estimator of dynamical systems (SEDS), Probabilistic Movement Primitives (ProMP) and Optimal Control Primitives (OCP). We compare these frameworks with respect to their movement encoding efficiency, reconstruction accuracy, and movement generalization capabilities. The new dataset consists of nine object manipulation actions performed by 12 humans: pick and place, put on top/take down, put inside/take out, hide/uncover, and push/pull with a total of 7,652 movement examples. Our analysis shows that for movement encoding and reconstruction DMPs and OCPs are the most efficient with respect to the number of parameters and reconstruction accuracy, if a sufficient number of kernels is used. In case of movement generalization to new start- and end-point situations, DMPs, OCPs and task parameterized GMM (TP-GMM, movement generalization framework based on tbGMR) lead to similar performance, which ProMPs only achieve when using many demonstrations for learning. All models outperform SEDS, which additionally proves to be difficult to fit. Furthermore we observe that TP-GMM and SEDS suffer from problems reaching the end-points of generalizations. These different quantitative results will help selecting the most appropriate models and designing trajectory representations in an improved task-dependent way in future robotic applications. • Comparison of five motion encoding frameworks on human manipulation action dataset. • Large dataset containing 7652 movement trajectories from 9 different action classes. • Comparison regarding encoding and generalization reveals model specific differences.

This paper proposes an enhanced robot skill learning system considering both motion generation and trajectory tracking. During robot learning demonstrations, dynamic movement primitives (DMPs) are used to model robotic motion. Each DMP consists of a set of dynamic systems that enhances the stability of the generated motion toward the goal. A Gaussian mixture model and Gaussian mixture regression are integrated to improve the learning performance of the DMP, such that more features of the skill can be extracted from multiple demonstrations. The motion generated from the learned model can be scaled in space and time. Besides, a neural-network-based controller is designed for the robot to track the trajectories generated from the motion model. In this controller, a radial basis function neural network is used to compensate for the effect caused by the dynamic environments. The experiments have been performed using a Baxter robot and the results have confirmed the validity of the proposed methods.