Primitive Library Research Articles

Research on LFD System of Humanoid Dual-Arm Robot

Although robots have been widely used in a variety of fields, the idea of enabling them to perform multiple tasks in the same way that humans do remains a difficulty. To solve this, we investigate the learning from demonstration (LFD) system with our independently designed symmetrical humanoid dual-arm robot. We present a novel action feature matching algorithm. This algorithm accurately transforms human demonstration data into task models that robots can directly execute, considerably improving LFD’s generalization capabilities. In our studies, we used motion capture cameras to capture human demonstration actions, which included combinations of simple actions (the action layer) and a succession of complicated operational tasks (the task layer). For the action layer data, we employed Gaussian mixture models (GMM) for processing and constructing an action primitive library. As for the task layer data, we created a “keyframe” segmentation method to transform this data into a series of action primitives and build another action primitive library. Guided by our algorithm, the robot successfully imitated complex human tasks. Results show its excellent task learning and execution, providing an effective solution for robots to learn from human demonstrations and significantly advancing robot technology.

Coordinated Motion Planning for Heterogeneous Autonomous Vehicles Based on Driving Behavior Primitives

Heterogeneous autonomous vehicle (HAV) coordinated motion planning must guide each vehicle out of the conflict zone based on the differences in vehicle platform characteristics. Decomposing complex driving tasks into primitives is an effective way to improve algorithm efficiency. Hence, the purpose of this paper is to complete the coordinated motion planning tasks through offline driving behavior primitive (DBP) library generation, online extension and selection of DBPs. The proposed algorithm applies dynamic movement primitives and singular value decomposition to learn driving behavior patterns from driving data, integrates them into a model-based optimization generation method as constraints, and builds a DBP library by fusing driving data and vehicle model. Based on the generated DBP library and primitive association probabilities learned from labeled driving segments via stochastic context-free grammar, the planning method completes the independent DBP extension of each vehicle in the conflict zone, generates an interaction DBP tree, and uses the mixed-integer linear programming algorithm to optimally select the primitives to be executed. This study demonstrates that the generated DBP library not only expands the types of primitives, but also distinguishes the characteristics of HAVs. We also present how to utilize the DBP libraries to obtain coordinated motion planning results with spatiotemporal information in the form of DBP extension and selection. The results obtained by real vehicle platforms and simulation show that the proposed method can accomplish coordinated motion planning tasks without relying on specific scene elements and highlight the unique motion characteristics of HAVs.

Selecting Minimal Motion Primitive Libraries with Genetic Algorithms

Motion primitives allow for application of discrete search algorithms to rapidly produce trajectories in complex continuous space. The maneuver automaton (MA) provides an elegant formulation for creating a primitive library based on trims and maneuvers. However, performance is fundamentally limited by the contents of the primitive library. If the library is too sparse, performance can be poor in terms of path cost, whereas a library that is too large can increase run time. This work outlines new methods for using genetic algorithms to prune a primitive library. The proposed methods balance the path cost and planning time while maintaining the reachability of the MA. The genetic algorithm in this paper evaluates and mutates populations of motion primitive libraries to optimize both objectives. We illustrate the performance of these methods with a simulated study using a nonlinear medium-fidelity F-16 model. We optimize a library with the presented algorithm for obstacle-free navigation and a nap-of-the-Earth navigation task. In the obstacle-free navigation task, we show a tradeoff of a 10.16% higher planning cost for a 96.63% improvement in run time. In the nap-of-the-Earth task, we show a tradeoff of a 9.712% higher planning cost for a 92.06% improvement in run time.

Utilizing Reinforcement Learning to Continuously Improve a Primitive-Based Motion Planner

This paper describes how the performance of motion primitive-based planning algorithms can be improved using reinforcement learning. Specifically, we describe and evaluate a framework that autonomously improves the performance of a primitive-based motion planner. The improvement process consists of three phases: exploration, extraction, and reward updates. This process can be iterated continuously to provide successive improvement. The exploration step generates new trajectories, and the extraction step identifies new primitives from these trajectories. These primitives are then used to update rewards for continued exploration. This framework required novel shaping rewards, development of a primitive extraction algorithm, and modification of the Hybrid A* algorithm. The framework is tested on a navigation task using a nonlinear F-16 model. The framework autonomously added 91 motion primitives to the primitive library and reduced average path cost by 21.6 s, or 35.75% of the original cost. The learned primitives are applied to an obstacle field navigation task, which was not used in training, and reduced path cost by 16.3 s, or 24.1%. Additionally, two heuristics for the modified Hybrid A* algorithm are designed to improve effective branching factor.

Tunnel Reconstruction With Block Level Precision by Combining Data-Driven Segmentation and Model-Driven Assembly

Metro subway systems with underground tunnels form the backbone of urban transportations and therefore, accurate monitoring and maintenance of such subway systems are extremely necessary for a hassle-free daily commutation of billions of people. Though 3-D models of tunnels are widely used for the deformation monitoring of such subway tunnels, existing model-based tunnel monitoring systems rely on coarse geometric models and hence fail to capture complete tunnel health information. We present a two-stage algorithm to create high-fidelity geometric models of tunnel lining from Terrestrial Laser Scanning (TLS) point clouds. Tunnel geometry, defined at the detailed block entity level, is constructed through a data-driven block segmentation algorithm and a model-driven assembly technique. In our approach, the 3-D tunnel block segmentation problem has been translated into a bolt and lining joint recognition problem from 2-D images unfolded from the 3-D scans. The segmented 3-D blocks are matched with a set of predefined 3-D templates from a primitive library via a constraint total least squares matching method and the matched 3-D templates are assembled to create the final watertight tunnel model. The proposed tunnel modeling method has been comprehensively evaluated on Changzhou, Nanjing, and Wuhan tunnel data sets in terms of outliers, missing data, point density, topological representation, robustness, and geometric accuracy. The experiments on Nanjing and Changzhou metro tunnels show that the geometric model fitting incurs an error of only 7 mm, which is almost consistent with a mean density of 6 mm of these two data sets. Experimental results validate the advantages and potentials of the proposed tunnel modeling method.

A VecGeom navigator plugin for Geant4

VecGeom is a geometry modeller library with hit-detection features as needed by particle detector simulation at the LHC and beyond. It was incubated by a Geant-R&D initiative and the motivation to combine the code of Geant4 and ROOT/TGeo into a single, better maintainable piece of software within the EU-AIDA program. So far, VecGeom is mainly used by LHC experiments as a geometry primitive library called from Geant4, where it was shown to provide 7–12% reduction in CPU time due to its faster algorithms for complex primitives [1]. In this contribution, we discuss how VecGeom can be used as the navigating library in Geant4 in order to benefit from both its fast geometry primitives as well as its vectorised navigation module. We investigate whether this integration provides the speed improvements expected, in addition to the gain obtained from geometry primitives. We discuss and benchmark the application of a VecGeomnavigator plugin to Geant4 for a simplified geometry and show paths towards production usage.

Learning motion primitives for planning swift maneuvers of quadrotor

This work proposes a novel, learning-based method to leverage navigation time performance of unmanned aerial vehicles in dense environments by planning swift maneuvers using motion primitives. In the proposed planning framework, desirable motion primitives are explored by reinforcement learning. Two-stage training composed of learning in simulations and real flights is conducted to build up a swift motion primitive library. The library is then referred in real-time and the primitives are utilized by an intelligent control authority switch mechanism when swift maneuvers are needed for particular portions of a trajectory. Since the library is constructed upon realistic Gazebo simulations and real flights together, probable modeling uncertainties which can degrade planning performance are minimal. Moreover, since the library is in the form of motion primitives, it is computationally inexpensive to be retained and used for planning as compared to solving optimal motion planning problem algebraically. Overall, the proposed method allows for exceptional, swift maneuvers and enhances navigation time performance in dense environments up to 20% as being demonstrated by real flights with Diatone FPV250 quadrotor equipped with PX4 FMU.

Pattern Primitive Library Construction and Feature Analysis of Kirgiz Textile Pattern

Textile patterns in each minority which contain cultural, historical, artistical, and reginal features are precious because of its unique aesthetic characteristics. More than 78 typical classical patterns of Kirgiz Yormac which is a kind of Kirgiz home-made textile are collected through field investigation, and digitalized by CorelDraw based on the concept of pattern primitives. Three methods for obtaining the pattern primitive units are proposed, and a pattern primitive library is established. It is found that the most popular Yormac pattern items are the ram horn, the flying eagles, and geometric figures, with a symmetrical configuration and structure. Red is the most epidemic color with the proportion over 70% of the pattern collected. By statistically and literature analysis, it can be concluded that Kirgiz textile pattern comes from life, being used in life. The custom, implication, spirit of Kirgiz shown in pattern come down in one continuous line with national pattern, and are highly in conformity with the culture and tradition of Chinese nation. The feature analysis of Yormac pattern, and construction of pattern primitive library is done for better comprehension of Kirgiz culture, and thus protecting and inheriting the Kirgiz intangible culture heritage.

Probabilistic Prioritization of Movement Primitives

Movement prioritization is a common approach to combine controllers of different tasks for redundant robots, where each task is assigned a priority. The priorities of the tasks are often hand-tuned or the result of an optimization, but seldomly learned from data. This paper combines Bayesian task prioritization with probabilistic movement primitives to prioritize full motion sequences that are learned from demonstrations. Probabilistic movement primitives (ProMPs) can encode distributions of movements over full motion sequences and provide control laws to exactly follow these distributions. The probabilistic formulation allows for a natural application of Bayesian task prioritization. We extend the ProMP controllers with an additional feedback component that accounts inaccuracies in following the distribution and allows for a more robust prioritization of primitives. We demonstrate how the task priorities can be obtained from imitation learning and how different primitives can be combined to solve even unseen task-combinations. Due to the prioritization, our approach can efficiently learn a combination of tasks without requiring individual models per task combination. Further, our approach can adapt an existing primitive library by prioritizing additional controllers, for example, for implementing obstacle avoidance. Hence, the need of retraining the whole library is avoided in many cases. We evaluate our approach on reaching movements under constraints with redundant simulated planar robots and two physical robot platforms, the humanoid robot “iCub” and a KUKA LWR robot arm.

Toward a flexible framework for learning: F3L

For stable and adaptable movement generation, it is required that movement primitives are generalizable in order to be executed in different contexts and for distinct instances of the same task. Therefore, a motion primitive must adapt policy parameters according to environment variables, instead of repetitively executing the exact same motor commands when placed into a specific context. Another important aspect is how the learning process of motion primitives is guided. Some movements are too complex to allow its primitives to be learned all at once, i.e, learning all their intricacies without a properly structured approach. In order to simplify this process, a framework by the name of Flexible Framework for Learning (F3L), that allows storing, combining and mutual and partial learning of parameterized policies, is presented. It provides the means, for instance, to build a movement primitive library or to perform offline gradual learning of a specific motor skill. Additionally, F3L is used to teach the controller of a biped robot (DARwIn-OP) to adapt the parameters of CPG-based locomotion according to the slope angles of the underlying surface. The introduction of several novel feedback mechanisms is also suggested. The results achieved are gratifying, promising and attest the fact that F3L can provide an effective and useful tool to structure the learning process of complex tasks. The direction correction feedback mechanism, in addition to keeping the robot walking in a straight path, has shown the ability to control walking direction bypassing internal CPG dynamics.

Learning movement primitive libraries through probabilistic segmentation

Movement primitives are a well-established approach for encoding and executing movements. While the primitives themselves have been extensively researched, the concept of movement primitive libraries has not received similar attention. Libraries of movement primitives represent the skill set of an agent. Primitives can be queried and sequenced in order to solve specific tasks. The goal of this work is to segment unlabeled demonstrations into a representative set of primitives. Our proposed method differs from current approaches by taking advantage of the often neglected, mutual dependencies between the segments contained in the demonstrations and the primitives to be encoded. By exploiting this mutual dependency, we show that we can improve both the segmentation and the movement primitive library. Based on probabilistic inference our novel approach segments the demonstrations while learning a probabilistic representation of movement primitives. We demonstrate our method on two real robot applications. First, the robot segments sequences of different letters into a library, explaining the observed trajectories. Second, the robot segments demonstrations of a chair assembly task into a movement primitive library. The library is subsequently used to assemble the chair in an order not present in the demonstrations.

Learning and Chaining of Motor Primitives for Goal-directed Locomotion of a Snakelike Robot with Screw-drive Units

Motor primitives provide a modular organization to complex behaviours in both vertebrates and invertebrates. Inspired by this, here we generate motor primitives for a complex snake-like robot with screw-drive units, and thence chain and combine them, in order to provide a versatile, goal-directed locomotion for the robot. The behavioural primitives of the robot are generated using a reinforcement learning approach called “Policy Improvement with Path Integrals” (PI2). PI2 is numerically simple and has the ability to deal with high-dimensional systems. Here, PI2 is used to learn the robot's motor controls by finding proper locomotion control parameters, like joint angles and screw-drive unit velocities, in a coordinated manner for different goals. Thus, it is able to generate a large repertoire of motor primitives, which are selectively stored to form a primitive library. The learning process was performed using a simulated robot and the learned parameters were successfully transferred to the real robot. B...

Robot learns from human teacher through modified kinesthetic teaching

Abstract Teaching new motor tasks to robots through physical interactions is an important goal for both robotics and machine learning. Most monolithic machine learning approaches fail to scale when going beyond basic skills. In this paper we present a simple framework for teaching the robot (to play tennis) through direct physical interaction with a human teacher (i.e. Kinesthetic Teaching). Current popular established method of kinesthetic teaching generally uses a two-stage approach: First, a library of motor primitives is generated through direct physical manipulation of the robot. In second stage, a reinforced learning (“reward” stage) is implemented to dynamically adjust the policy of choosing from motor primitive library. In this paper, we show that by proper modification of the first stage of Kinesthetic Teaching and incorporating the domain experience of the human teacher, we can remove the necessity of the second stage. This approach has multiple advantages: (i) We can make the whole training process much simpler. This would go a long way in making our training algorithm scalable (for much increased number of basic moves, etc). (ii) One potential problem with the “reward” learning phase is that there may be subjective difference of what is a “good” shot from the kinesthetic teaching and from the bystander viewpoint (later in “reward” stage). Even a little difference in this regard will result in a confusing feedback (“reward”), and hence it would be difficult to correctly figure out which library training samples should be reassigned what weight values. Our approach eliminates this problem altogether.

A hierarchical approach for primitive-based motion planning and control of autonomous vehicles

A hierarchical approach for motion planning and control of nonlinear systems operating in obstacle environments is presented. To reduce the computation time during the motion planning process, dynamically feasible trajectories are generated in real-time through concatenation of pre-specified motion primitives. The motion planning task is posed as a search over a directed graph, and the applicability of informed graph search techniques is investigated. Specifically, we develop a locally greedy algorithm with effective backtracking ability and compare this algorithm to weighted A* search. The greedy algorithm shows an advantage with respect to solution cost and computation time when larger motion primitive libraries that do not operate on a regular state lattice are utilized. Linearization of the nonlinear system equations about the motion primitive library results in a hybrid linear time-varying model, and an optimal control algorithm using the ℓ2-induced norm as the performance measure is provided to ensure that the system tracks the desired trajectory. The ability of the resulting controller to closely track the trajectory obtained from the motion planner, despite various disturbances and uncertainties, is demonstrated through simulation.

Effects of Building Primitive Library and Automatic Correction of Roof Topology Graph

Abstract. The reconstruction of 3D building models has been extensively researched in the last decade. The model driven methods, which fit building parts to predefined building primitives and glue them together into complete buildings, have been proved to be powerful. This paper solves two open questions left in prior works. Firstly, the primitive building libraries defined in literature are limited to few types of building structures. Therefore the undefined types of buildings could not be well represented by the primitive libraries. Secondly, the roof topology graphs used for interpreting the roof structures are assumed to be correct in previous works. In many cases, however, the roof topology graphs are inevitably wrong because of outliers in input data and errors in their construction steps. In this paper we solve the two problems and compare the improvements with prior woks. We introduce a complete building library, which can present planer buildings with arbitrary structures, and present an automatic method to correct the erroneous roof topology graph. Experiments show our method is robust to reconstruct buildings in both resident and commercial areas.

Postures and Motions Library Development for Verification of Ground Crew Human Factors Requirements

Spacecraft and launch vehicle ground processing activities require a variety of unique human activities. These activities are being documented in a primitive motion capture library. The library will be used by human factors engineering analysts to infuse real to life human activities into the CAD models to verify ground systems human factors requirements. As the primitive models are being developed for the library, the project has selected several current human factors issues to be addressed for the Space Launch System (SLS) and Orion launch systems. This paper explains how the motion capture of unique ground systems activities is being used to verify the human factors engineering requirements for ground systems used to process the SLS and Orion vehicles, and how the primitive models will be applied to future spacecraft and launch vehicle processing.

Design and Implementation of an SVG Editor for Power System

It is the fundamental requirement for power system graphical editor to be able to draw rapidly, precisely and allocate corresponding configuration parameters for the power components and reserve operation interface. Element of this editor uses the SVG format and defines metadata tags based on the CIM specification carrying metadata in the SVG document which makes the graphics and data can be linked to a unified canonical form. The editor achieves the purpose as follows: editing, storing, and exporting a wiring diagram; importing a wiring diagram from other systems; automatically extracting and storing layers and elements in the importing process; creating and maintaining the user-defined primitive library or layer library and binding associated data. This SVG graphics editor for power system has met the functional requirements of the user and has been applied in several practical projects.

A generative statistical approach to automatic 3D building roof reconstruction from laser scanning data

This paper presents a generative statistical approach to automatic 3D building roof reconstruction from airborne laser scanning point clouds. In previous works, bottom-up methods, e.g., points clustering, plane detection, and contour extraction, are widely used. Due to the data artefacts caused by tree clutter, reflection from windows, water features, etc., the bottom-up reconstruction in urban areas may suffer from a number of incomplete or irregular roof parts. Manually given geometric constraints are usually needed to ensure plausible results. In this work we propose an automatic process with emphasis on top-down approaches. The input point cloud is firstly pre-segmented into subzones containing a limited number of buildings to reduce the computational complexity for large urban scenes. For the building extraction and reconstruction in the subzones we propose a pure top-down statistical scheme, in which the bottom-up efforts or additional data like building footprints are no more required. Based on a predefined primitive library we conduct a generative modeling to reconstruct roof models that fit the data. Primitives are assembled into an entire roof with given rules of combination and merging. Overlaps of primitives are allowed in the assembly. The selection of roof primitives, as well as the sampling of their parameters, is driven by a variant of Markov Chain Monte Carlo technique with specified jump mechanism. Experiments are performed on data-sets of different building types (from simple houses, high-rise buildings to combined building groups) and resolutions. The results show robustness despite the data artefacts mentioned above and plausibility in reconstruction.

Comparison of the Frequency of Functional SH3 Domains with Different Limited Sets of Amino Acids Using mRNA Display

Although modern proteins consist of 20 different amino acids, it has been proposed that primordial proteins consisted of a small set of amino acids, and additional amino acids have gradually been recruited into the genetic code. This hypothesis has recently been supported by comparative genome sequence analysis, but no direct experimental approach has been reported. Here, we utilized a novel experimental approach to test a hypothesis that native-like globular proteins might be easily simplified by a set of putative primitive amino acids with retention of its structure and function than by a set of putative new amino acids. We performed in vitro selection of a functional SH3 domain as a model from partially randomized libraries with different sets of amino acids using mRNA display. Consequently, a library rich in putative primitive amino acids included a larger number of functional SH3 sequences than a library rich in putative new amino acids. Further, the functional SH3 sequences were enriched from the primitive library slightly earlier than from a randomized library with the full set of amino acids, while the function and structure of the selected SH3 proteins with the primitive alphabet were comparable with those from the 20 amino acid alphabet. Application of this approach to various combinations of codons in protein sequences may be useful not only for clarifying the precise order of the amino acid expansion in the early stages of protein evolution but also for efficiently creating novel functional proteins in the laboratory.

OpenCL and parallel primitives for digital TV applications

Open Computing Language®(OpenCL®), which is created to support H. Yeo parallel programming of heterogeneous multicore-processor systems, has a very large potential for high-performance computing and consumer electronics since it provides application programming interfaces (APIs) to help make a portable code that runs across multiple devices. OpenCL is still under development, and it is not clear whether OpenCL has any advantages over other frameworks aside from portability. The purpose of our project was to define evaluation criteria, empirically evaluate OpenCL as a programming framework using evaluation criteria (e.g., performance, productivity, and portability criteria), define and implement parallel primitives in OpenCL, and demonstrate how the use of the implemented parallel primitives can have benefits for our target applications. Parallel primitive library APIs are defined to implement parallel algorithms in OpenCL, and a set of data- and task-parallel primitives is implemented and incorporated in the target applications. Multicore central processing units, the Cell Broadband Engine®(Cell/B.E.®), and graphics processing units are used as target platforms, and digital TV applications are used to evaluate usefulness of OpenCL. Preliminary results show that parallel primitives can be one of the ways to improve application performance and programmer productivity with respect to OpenCL while still maintaining software portability.