Robot Tool Path Research Articles

DeepCraft: imitation learning method in a cointelligent design to production process to deliver architectural scenarios

The recent advancements in digital technologies and artificial intelligence in the architecture, engineering, construction, and operation sector (AECO) have induced high demands on the digital skills of human experts, builders, and workers. At the same time, to satisfy the standards of the production-efficient AECO sector by reducing costs, energy, health risk, material resources, and labor demand through efficient production and construction methods such as design for manufacture and assembly (DfMA), it is necessary to resolve efficiency-related problems in mutual human‒machine collaborations. In this article, a method utilizing artificial intelligence (AI), namely, generative adversarial imitation learning (GAIL), is presented then evaluated in two independent experiments related to the processes of DfMA as an efficient human‒machine collaboration. These experiments include a) training the digital twin of a robot to execute a robotic toolpath according to human gestures and b) the generation of a spatial configuration driven by a human's design intent provided in a demonstration. The framework encompasses human intelligence and creativity, which the AI agent in the learning process observes, understands, learns, and imitates. For both experimental cases, the human demonstration, the agent's training, the toolpath execution, and the assembly configuration process are conducted digitally. Following the scenario generated by an AI agent in a digital space, physical assembly is undertaken by human builders as the next step. The implemented workflow successfully delivers the learned toolpath and scalable spatial assemblies, articulating human intelligence, intuition, and creativity in the cocreative design.

Fiber Reinforcement of 3D Printed Concrete by Material Extrusion Toolpaths Aligned to Principal Stress Trajectories

Additive manufacturing of cementitious materials is a rapidly growing branch of manufacturing both in research and industry, particularly the variant of material deposition by extrusion. This process results in a strong anisotropy in mechanical properties, owing largely to the interfaces between adjacent filaments. This anisotropy is even more pronounced when fiber reinforced mortars or continuous entrained reinforcement components such as cables are used. To exploit orientation-dependent performance, the print path can be designed to align with the principal (tensile) stress trajectories. However, obtaining an appropriate print path based on this concept poses several challenges, related to the filling of intermediate spaces between two trajectories. In this paper, an approach for planning such a robot toolpath is presented, elaborated, and illustrated by means of a case study on a well-known reference case. The main features of the tool planning method are the relaxation of the offset width, the avoidance of toolpaths with acute angles by intersecting offset curves, and a continuous toolpath.

MULTISENSOR FUSION-BASED DIGITAL TWIN IN ADDITIVE MANUFACTURING FOR IN-SITU QUALITY MONITORING AND DEFECT CORRECTION

AbstractEarly detection and correction of defects are critical in additive manufacturing (AM) to avoid build failures. In this paper, we present a multisensor fusion-based digital twin for in-situ quality monitoring and defect correction in a robotic laser-directed energy deposition process. Multisensor fusion sources consist of an acoustic sensor, an infrared thermal camera, a coaxial vision camera, and a laser line scanner. The key novelty and contribution of this work are to develop a spatiotemporal data fusion method that synchronizes and registers the multisensor features within the part's 3D volume. The fused dataset can be used to predict location-specific quality using machine learning. On-the-fly identification of regions requiring material addition or removal is feasible. Robot toolpath and auto-tuned process parameters are generated for defect correction. In contrast to traditional single-sensor-based monitoring, multisensor fusion allows for a more in-depth understanding of underlying process physics, such as pore formation and laser-material interactions. The proposed methods pave the way for self-adaptation AM with higher efficiency, less waste, and cleaner production.

Parametric modeling approach in laser wire additive manufacturing process

In recent years, the evolution of computer-aided design technology has established a new design method to improve digital model creations. The new design method is called Parametric Modeling. In this paper, 3D metallic models are printed using a novel Parametric Modeling approach. The goal of this approach is to use parametric design features to simulate and print 3D metallic objects using Laser Wire Additive Manufacturing (LWAM) process. The proposed approach includes a pattern creation and robot targets assignment while considering several process requirements of LWAM and the robot system. This technique will allow the development of an adaptive robot toolpath for a good deposition process. Finally, a wall, a cylinder, and a complex shape were simulated and deposited to validate the proposed approach. The results show that the approach is feasible, adaptive, and can enhance 3D metallic print in the LWAM process.

Uniform Coverage Tool Path Generation for Robotic Surface Finishing of Curved Surfaces

We provide a novel solution to the problem of robot tool path generation to uniformly cover the surface area of a curved surface with varying curvatures by using a circular, flat finishing tool. We first develop a local contact shape descriptor using tools from differential geometry to estimate the contact shape and its area between the finishing tool and the workpiece surface under an applied normal force. Based on this varying contact area as the tool moves on the surface, we describe a tool path planning strategy that provides full and uniform coverage of a free-form or curved surface with the least amount of overlap. As opposed to fixed spacing paths with point contact in most existing approaches, we provide a novel path generation method where the interval spacing between neighboring path segments is determined by the contact area, which is position-dependent and varies along the surface in accordance with its principal curvatures. An efficient bisection method is employed in the proposed off-line path planning strategy. Numerical studies were conducted for ruled and general curved surfaces to compare the contact areas and coverage of the resulting path with paths generated using two other approaches from the literature. Real-time surface finishing experiments with a six degree-of-freedom robot equipped with a surface finishing tool were also conducted for both these surfaces to validate the proposed approach. Results from the numerical studies and experiments are presented and discussed.

Development of near homogeneous properties in wire arc additive manufacturing process for near-net shaped structural component of low-carbon steel

Low-carbon steel is a common structural material, but additively manufactured structural component of this material is rare due to its inhomogeneous properties. In this article, the wire arc additive manufacturing method was used to achieve near homogeneous properties of a low-carbon steel structural component. The process heat input was optimised for the desired layer geometry, and then the optimal energy was applied with a time delay to deposit individual layers. The time delay was used to achieve cyclic heating and cooling treatment of deposited layers. The best possible robotic tool path movement with multi point arcing was further adopted in the study to achieve proper thermal distribution across the structural component. The microstructure of layers was dominated by quasi-polygonal ferrite morphology and pearlite precipitation, with little variation in quantity across the component. The hardness profile was almost consistent with the average hardness of ∼176.92 HV. The proof stress slightly increases with decrease in grain size and increase in ferrite/pearlite ratio, however, the overall tensile behaviour is homogeneous with average σ0.2, σu and ε% values of 427.78 MPa, 527.89 MPa and 22.31%, respectively. The quasi-ductile fracture was generally occurred due to void coalescence around larger inclusions. The overall analysis showed that more than 90% of homogeneity was achieved in microstructural and mechanical behaviour of the deposited component.

Robot Operations for Pine Tree Resin Collection

A robotic technology consisting of an industrial robot mounted on an autonomous rover used to tap slash pine trees and collect their oleoresin for processing is introduced, and the technological challenges related to the robotic operations are discussed in detail. Unlike the case of industrial automated manufacturing systems where the relative position between the tool and workpiece can be controlled within a few hundredths of a millimeter accuracy, when used in highly unstructured environments characteristic to forestry or agriculture, the positioning accuracy between the industrial robot and the target on which it operates can be much lower than the accuracy required for the operation of the industrial robot. The paper focuses on presenting the robotic operations necessary for drilling three converging boreholes in the pine tree, spraying the boreholes with chemicals, inserting a plastic tube with pre-attached collection bag in one borehole and inserting two plugs in other two boreholes. The challenges related to performing these robotic operations in conditions of large variations in the actual shape of the pine tree trunk and variations in the relative position between the robot and the pine tree after the autonomous vehicle positions itself in front of the tree are presented. The technical solutions used to address these challenges are also described. The strategies used to programmatically adjust the robot toolpath based on detection of the borehole entry points and on the measurement of the insertion force are presented.

Robotic spray coating of self-sensing metakaolin geopolymer for concrete monitoring

Sensors and new materials can support optimised concrete maintenance, and produce the data needed to justify new, low carbon structural designs. While these technologies are affordable, the process of manual installation in a construction context comes with acute and unfamiliar risks to productivity, personnel safety, and confidence in the quality of workmanship. The installation of smart materials using robotics could address some of these issues, but there are few proofs-of-concept at the time of writing. Here, we present a robotically controlled process for spray coating geopolymers — a class of self-sensing concrete repair materials. By tuning mix design, robotic toolpaths and spray dispenser parameters, we show reliable and automated spray coating of 250 mm2 patches in a laboratory setting. The cured geopolymer has a compressive strength of 20 MPa, and a bond strength to the concrete substrate of 0.5 MPa. Electrical interrogation of patches, via a set of four electrodes, produces strain and temperature measurements of the underlying concrete substrate with resolutions of 1 με and 0.2 °C, respectively. This demonstration multifunctional material deposition using robotics is a step towards remote, traceable, and low-risk technology deployment across civil engineering sectors. This could support more widespread adoption of novel concrete health monitoring and repair systems in future.

An experimental investigation of the effectiveness of Ar-CO2 shielding gas mixture for the wire arc additive process

Wire arc additive manufacturing (AM) is the process by which a large, metallic structure is built layer-by-layer using a welding arc to melt a wire feedstock. A novel opportunity exists to alter the shielding gas composition to fabricate distinct geometrical features without altering the other AM parameters. In the following study, shielding gases with varied concentrations of Argon (Ar) and CO2 was used to deposit three distinct geometric shapes (walls, infill, and overhang) using a wire-based additive manufacturing system utilizing the gas metal arc welding (GMAW) surface tension transfer (STT) process. Computer-aided design (CAD) models were sliced with a custom-built slicer, and the sliced algorithm was converted into optimal robotic toolpaths. A custom virtual instrument (VI) was built in LabVIEW to compare the temperature profiles on the surface during each deposition process. After each deposition, the geometric features were scanned, and the surface waviness value was evaluated. Tensile and Charpy impact coupons were extracted from the wall geometries in the longitudinal and transverse directions and tested. The results indicated that a higher CO2 content produced higher melt pool temperatures to an extent, while lower contents of CO2 resulted in a dimensionally accurate geometry. The data also indicated that the 2%/98% CO2/Ar blend produced scatter in tensile strength and the analysis of variance (ANOVA) shows significant difference. However, the intermediate range of CO2 (5–10%) resulted in uniform tensile properties. Altogether, these results indicate that a 5%/95% CO2/Ar blend is the ideal shielding gas for lowering process temperatures and improving mechanical properties in wire arc additive manufacturing using the gas metal arc welding surface tension transfer process. Additionally, varying concentrations of Ar/CO2 can be used within the same part in order to modify the local properties or process parameters such as strength, toughness, temperature, or dimensional features. This may improve overall manufacturing quality without sacrificing specific properties.

Additive manufacturing of cantilever - From masonry to concrete 3D printing

3d printing of cementitious material is a relatively new additive manufacturing process whose growing interest and fast development is mainly due to the digitalised manufacturing, allowing the disposition of material where it pleases. Yet, due to the properties of the fresh material and the difficulty to generate paths for the robots, the printed geometries have remained simple. In this regard, this papers longs to broaden the range of printable shapes by proposing a process-aware exploration of the 3d printing design space.This is done by looking at historic strategies that have been developed to build cantilevers, vaults and domes in masonry - a more ancient additive manufacturing process. Similarities and main differences between the two processes are pointed out, at the scale of the component, the layer and the global structure. From that a classification of masonry strategies to build cantilevers is proposed, facilitating the identification of parameters for 3d printing that will allow to reproduce such structures. Later, some guidelines for the design of printable geometries and the generation of robotic toolpaths are given, in the light of previous findings.

Optimization and prefabrication of timber Voronoi shells

We introduce a variant of the timber grid shell, the timber Voronoi shell, whose surface is reticulated by Voronoi tessellation and whose edges are made of discrete dimensional timber. This work explores form-finding methods of the Voronoi shell as a compressive funicular shell. Two closed-form solutions to shape initialization are proposed. We develop methods for minimizing the deviation from coplanarity between timber members and adjacent surface normal in order to facilitate manufacturing. A well-defined fabrication process is important for making the physical structure consistent with the structural model. A 6-axis robot with a motor spindle is employed to prefabricate the timber so the in situ manual assembly becomes easier. A parametric model describes the joint details. We formulate the robotic toolpath as a closed-form function of the resultant mesh from form finding. Thus, a general-purpose programming language can directly implement the mesh optimization and manufacturing processes without CAD or CAM software. The physical implementations, including an exhibition pavilion, validated the approach.

An Augmented Reality Framework for Robotic Tool-path Teaching

Robot manipulators are widely used in industries to automate myriad operations. To address the demand shifts in industries, automated systems such as robots must flexibly adapt in response. Such dynamics often necessitate end-users to reprogram robots to meet the changing industry needs. This leads to a demand for experienced programmers, familiar with proprietary robot software. As one alternative, an augmented reality (AR) framework can offer a user-friendly solution that permits end-users to reprogram robots without any domain expertise. This paper presents an AR teaching (ART) methodology that allows end-users to program varied manipulators in an intuitive and effortless manner for tool-path teaching. The ART method is contrasted with an alternative kinesthetic teaching method for its performance and user experience. Results show that the ART method yields a convenient and time-efficient teaching method and it is recommended by users over the kinesthetic teaching method.

Permanent Magnets Produced by Cold Spray Additive Manufacturing for Electric Engines

With the global trend towards electrification of transport, the interest for various configurations of electric motors is growing. Unfortunately, the low versatility of the motor magnet fabrication technologies limits the development of new motor geometries. Cold spray additive manufacturing was used for shaping permanent magnets for the direct fabrication of electric motor parts without the need for additional assembly steps. This novel technique allows an increase in the design flexibility of electrical machine geometries targeting improved performance. The permanent magnets were deposited from a NdFeB-Al composite powder mix featuring high raw theoretical remanence and coercivity. The role of the process parameters such as powder mix composition, binder granulometry and gas temperature on the magnetic volume fraction and the resulting magnetic properties will be investigated and compared with data from the literature. The use of complex robot toolpath programming necessary to spray on motor parts will be illustrated through an example of the fabrication of a complex shape rotor. The obtained magnetic properties demonstrate the feasibility of using cold spray additive manufacturing as an effective technology to fabricate motor parts without additional assembly steps.

3D 프린팅 제품의 후처리 가공을 위한 5자유도 로봇 기반 가공시스템 설계 및 가공경로 생성

최근 다양한 산업분야에서 3D 프린터를 이용하여 제작된 제품의 활용이 증가하고 있다. 하지만 3D 프린팅 제품은 공정의 특성 상 표면품질과 형상정밀도가 낮다. 따라서 3D 프린팅을 이용하여 정교한 제품을 제작하기 위해 후처리 가공이 필수적이며, 이에 소비되는 시간과 비용은 3D 프린팅 공정의 생산성을 저하시킨다. 본 연구에서는 3D 프린팅 제품의 후처리 가공을 위한 로봇가공시스템을 제안하였다. 5자유도 로봇팔에 스핀들을 부착한 가공시스템을 설계하고, 공구의 위치를 제어하기 위한 제어기를 제작하였다. 공구의 위치와 자세를 결정하여 가공 경로를 생성하는 알고리즘을 제작하고, 그 성능을 검증하였다. 3D 프린팅 제품의 후처리 가공 전과 후의 가공 표면 조도를 비교함으로써 제안한 방법의 성능을 검증하였다.

A Robotic Deburring Methodology for Tool Path Planning and Process Parameter Control of a Five-Degree-of-Freedom Robot Manipulator

Industrial robotics is a continuously developing domain, as industrial robots have demonstrated to possess benefits with regard to robotic automation solutions in the industrial automation field. In this article, a new robotic deburring methodology for tool path planning and process parameter control is presented for a newly developed five-degree-of-freedom hybrid robot manipulator. A hybrid robot manipulator with dexterous manipulation and two experimental platforms of robot manipulators are presented. A robotic deburring tool path planning method is proposed for the robotic deburring tool position and orientation planning and the robotic layered deburring planning. Also, a robotic deburring process parameter control method is proposed based on fuzzy control. Furthermore, a dexterous manipulation verification experiment is conducted to demonstrate the dexterous manipulation and the orientation reachability of the robot manipulator. Additionally, two robotic deburring experiments are conducted to verify the effectiveness of the two proposed methods and demonstrate the highly efficient and dexterous manipulation and deburring capacity of the robot manipulator.

Development of a Robotic Structural Steel Cutting System

Curved buildings and structures are getting popular and digital fabrication of the structural steels has been successfully tried in several ways. This study presents a development of a robotic system cutting structural steels and marking texts on the steels using digital information of shapes. A configuration of the overall system is explained based on functional decomposition of conventional cutting systems. The cutting system consists of a robotic system and a handling system, and the main functions of the cutting system are classified as translation, orientation, alignment, and cutting. A digital model of the system was created and tested in a robot simulation system. A robot tool-path generation system was also developed. The tool-path generation system uses the DSTV file, an industry standard for the shape of a structural steel, to generate robot job files. This study also provides robot workspaces in which the cutting tool can safely access and cut the workpieces. The proposed system was tested for cutting structural steels clamped in front of the robot. The results show that the generated tool-paths are valid in cutting the structural steel and the possibility of an overall robotic system for cutting structural steels.

Programming a Robot for Conformance Grinding of Complex Shapes by Capturing the Tacit Knowledge of a Skilled Operator

This paper describes a novel methodology to reduce the effort in automating manual surface finishing processes by bridging the knowledge transfer gap of the manual operator's skills to a robot program. Key process variables (KPVs), i.e., contact force, tool path, and feed rate, of the manual operator performing the task are captured with a “sensorized” hand-held belt grinder, while the changes to the work-piece geometry is captured using a 3-D scanner. The entire manual tool-path strategy is segmented into its primitives or primary strategies before programming an equivalent robotic tool-path and strategy. The manual tool-path primitives are imported into computer-aided-manufacturing software where boundary splines are created to generate the robotic tool-paths. An analytical material removal rate (MRR) model is used to scale the extracted manual KPVs such that the parameters can be executed by the robotic platform, while still maintaining an equivalent material removal profile. In the first experimental trial with the designed robotic finishing strategy using this approach, the work-piece could be finished to within 0.7 mm of the desired shape.

Introducing a novel mesh following technique for approximation-free robotic tool path trajectories

Abstract Modern tools for designing and manufacturing of large components with complex geometries allow more flexible production with reduced cycle times. This is achieved through a combination of traditional subtractive approaches and new additive manufacturing processes. The problem of generating optimum tool-paths to perform specific actions (e.g. part manufacturing or inspection) on curved surface samples, through numerical control machinery or robotic manipulators, will be increasingly encountered. Part variability often precludes using original design CAD data directly for toolpath generation (especially for composite materials), instead surface mapping software is often used to generate tessellated models. However, such models differ from precise analytical models and are often not suitable to be used in current commercially available path-planning software, since they require formats where the geometrical entities are mathematically represented thus introducing approximation errors which propagate into the generated toolpath. This work adopts a fundamentally different approach to such surface mapping and presents a novel Mesh Following Technique (MFT) for the generation of tool-paths directly from tessellated models. The technique does not introduce any approximation and allows smoother and more accurate surface following tool-paths to be generated. The background mathematics to the new MFT algorithm are introduced and the algorithm is validated by testing through an application example. Comparative metrology experiments were undertaken to assess the tracking performance of the MFT algorithms, compared to tool-paths generated through commercial software. It is shown that the MFT tool-paths produced 40% smaller errors and up to 66% lower dispersion around the mean values. Highlights This work presents a technique for generation of tool-paths from tessellated models. The technique does not introduce approximations to surface following tool-paths. Comparative metrology experiments were carried out to assess the technique. The technique produces smaller errors and lower dispersion around the mean values. The developed method is aligned with the growing use of surface mapping techniques.

Capturing the tacit knowledge of the skilled operator to program tool paths and tool orientations for robot belt grinding

This paper describes a novel methodology for generating grinding tool paths, orientations and grinding parameters based on the captured trajectories of a surface finishing tool operated by the skilled operator. First, a variable frame size moving window principal component analysis is performed on the trajectories. Next, features such as the amplitude, speed of the tool centre point, direction of travel, average position, contact force and orientation of the grinder are calculated. Kernel k-means clustering is applied to the trajectory’s feature vectors for segmentation into the tool path primitives. The robotic tool path primitives would be ordered similar to the manual tool path primitives in the grinding strategy. The robotic tool path primitives are then generated based on the boundaries of the manual tool path primitive in a computer-aided manufacturing software. Anchor points are selected from the corners and at intermediate sections along the robotic tool path to inherit the orientations of the manual tool path. Using the inherited skilled operator tool orientations, spherical linear interpolation and spherical spline quaternion interpolation are used to interpolate the orientations for the robotic tool path points along the cross curves, followed by the flow curves, respectively. A finite element model of the belt grinding process is also introduced to quantify the effect of the contact wheel serrations on the material removal rate. This is to account for the difference in contact wheel design between the handheld and robotic grinding tool. The programmed robotic tool paths generated from the trajectories captured from a skilled operator were proven to be able to grind the workpiece to the desired profile within the desired tolerance in a pre-production environment.

PAUT inspection of complex-shaped composite materials through six DOFs robotic manipulators

The requirement to increase inspection speeds for the non-destructive testing (NDT) of composite aerospace parts is common to many manufacturers. The prevalence of complex curved surfaces in the industry provides significant motivation for the use of six-axis robots for the deployment of NDT probes in these inspections. The IntACom project, developed by TWI Technology Centre (Wales) and supported by a number of major aerospace partners and the Welsh government, has produced a prototype robotic NDT system. The prototype system is capable of inspecting complex-geometry composite components with great time savings. Two six-axis robotic arms deploy end effectors carrying phased array ultrasonic testing (PAUT) probes. A simple-to-use graphical user interface (GUI) has been developed to control all aspects of the robotic inspection, from initial loading of part data, through scanning of the part to data analysis. The collaboration between TWI and the University of Strathclyde has boosted the establishment of new approaches for robotic tool-path generation, targeted to NDT inspections. Many unique features, such as the real-time B-scan for optimisation of PAUT settings and the external control of the robotic manipulators to allow returning to points of interest, increase the usefulness of the inspection process. This paper presents an overview of the project and of the research outcomes.