Alfred L. Wicks’s research while affiliated with Virginia Tech and other places

The design and organization of complex computational systems, such as those found in robots, traditionally requires laborious trial-and-error processes to ensure components are correctly connected with necessary resources for computation. This paper presents a novel generalization of the quadratic assignment and routing problem, introducing formalizations for selecting components and interconnections to synthesize a complete system capable of providing some user-defined functionality. By introducing problem context, functional requirements, and modularity directly into the assignment problem, optimal systems can be automatically generated while respecting constraints on bandwidth and computational resources. The ability to generate complete functional systems reduces manual design effort by allowing for a guided exploration of the design space, and increases resiliency by quantifying resource margins and enabling adaptation of system structure in response to changing environments, hardware or software failure. The proposed formulation is cast as an integer linear program which is provably $\mathcal{NP}$-hard. Several case studies explore the capabilities in designing several complex robot systems, highlighting the expressiveness and scale of problems addressable by this approach. Numerical simulations quantify real world performance and demonstrate tractable time complexity for the scale of problems encountered in many modern robotic systems.

In a previous study, we presented VT-Lane, a three-step framework for real-time vehicle detection, tracking, and turn movement classification at urban intersections. In this study, we present a case study incorporating the highly accurate trajectories and movement classification obtained via VT-Lane for the purpose of speed estimation and driver behavior calibration for traffic at urban intersections. First, we use a highly instrumented vehicle to verify the estimated speeds obtained from video inference. The results of the speed validation show that our method can estimate the average travel speed of detected vehicles in real-time with an error of 0.19 m/sec, which is equivalent to 2% of the average observed travel speeds in the intersection of the study. Instantaneous speeds (at the resolution of 30 Hz) were found to be estimated with an average error of 0.21 m/sec and 0.86 m/sec respectively for free-flowing and congested traffic conditions. We then use the estimated speeds to calibrate the parameters of a driver behavior model for the vehicles in the area of study. The results show that the calibrated model replicates the driving behavior with an average error of 0.45 m/sec, indicating the high potential for using this framework for automated, large-scale calibration of car-following models from roadside traffic video data, which can lead to substantial improvements in traffic modeling via microscopic simulation.

The layer-by-layer deposition process used in material extrusion (ME) additive manufacturing results in inter- and intra-layer bonds that reduce mechanical performance. Multi-axis ME techniques have shown potential for mitigating this issue by enabling tailored deposition directions based on loading conditions in three dimensions (3D). Planning deposition paths leveraging this capability remains a challenge, as an intelligent method for assigning these directions does not exist. Existing literature introduced topology optimization (TO) methods that assign material orientations to discrete regions of a part by simultaneously optimizing material distribution and orientation. These methods are insufficient for multi-axis ME, as the process offers additional freedom in varying material orientation that is not available to those methods. Additionally, optimizing orientation design spaces is difficult, and this issue is amplified with increased flexibility; the chosen orientation parameterization heavily impacts the algorithm's performance. Therefore, the authors i) present a TO method to solve the simultaneous problem with considerations for 3D material orientation variation and ii) establish a suitable parameterization of the orientation design space. Three parameterizations are explored in this work: Euler angles, explicit quaternions, and natural quaternions. The parameterizations are compared using two benchmark minimum compliance problems: a 2.5D Messerschimitt-Böolkow-Blohm beam and a 3D Wheel. For the Wheel, the presented algorithm demonstrated a 38% improvement in compliance over an algorithm that only allowed planar orientation variation. Additionally, natural quaternions maintain the well-shaped design space of explicit quaternions without unit length constraints, which lowers computational costs.

The thermal characteristics of the material extrusion additive manufacturing (AM) process produce weak bonds between layers and adjacent depositions, resulting in an overall anisotropic mechanical performance. Design for AM guidelines advise printing load-bearing parts such that load is applied strictly along the deposition paths, but this can be difficult to achieve with complex loading conditions. Recent works have explored toolpath generation techniques capable of generating deposition paths that are aligned with complicated load paths, but the methods rely on assumptions about the shape of the load paths relative to the geometry. In this paper, the authors present an algorithm for generating deposition paths for any arbitrary geometry and anticipated load paths. Deposition paths are planned using a streamline placement algorithm - commonly used for visualizing fluid flow fields - that treats the load paths as a velocity field. The algorithm is demonstrated on an example geometry, and the volumetric coverage of the resulting toolpath is compared to a toolpath generated using a standard toolpath planning technique. Through this comparative study, it is demonstrated that the toolpath resulting from the authors' proposed algorithm is able to follow the load paths while still achieving similar volumetric coverage to the standard toolpath.

Purpose Material extrusion (ME) suffers from anisotropic mechanical properties that stem from the three degree of freedom (DoF) toolpaths used for deposition. The formation of each layer is restricted to the XY-plane, which produces poorly bonded layer interfaces along the build direction. Multi-axis ME affords the opportunity to change the layering and deposition directions locally throughout a part, which could improve a part’s overall mechanical performance. The purpose of this paper is to evaluate the effects of changing the layering and deposition directions on the tensile mechanical properties of parts printed via multi-axis ME. Design/methodology/approach A multi-axis toolpath generation algorithm is presented and implemented on a 6-DoF robotic arm ME system to fabricate tensile specimens at different global orientations. Specifically, acrylonitrile butadiene styrene (ABS) tensile specimens are printed at various inclination angles using the multi-axis technique; the resulting tensile strengths of the multi-axis specimens are compared to similarly oriented specimens printed using a traditional 3-DoF method. Findings The multi-axis specimens had similar performances regardless of orientation and were equivalent to the 3-DoF specimens printed in the XYZ orientation (i.e. flat on the bed with roads aligned to the loading condition). This similarity is attributed to those sets of specimens having the same degree of road alignment. Practical implications Parts with out-of-plane loads currently require design compromises (e.g. additional material in critical areas). Multi-axis deposition strategies could enable local changes in layering and deposition directions to more optimally orient roads in critical areas of the part. Originality/value Though multi-axis ME systems have been demonstrated in literature, no prior work has been done to determine the effects of the deposition angle on the resulting mechanical properties. This work demonstrates that identical mechanical properties can be obtained irrespective of the build direction through multi-axis deposition. For ABS, the yield tensile strength of vertically oriented tensile bars was improved by 153 per cent using multi-axis deposition as compared to geometrically similar samples fabricated via 3-DoF deposition.

Due to the layer stacking inherent in traditional three-axis material extrusion (ME) additive manufacturing processes, a part's mechanical strength is limited in the print direction due to weaker interlayer bond strength. Often, this requires compromise in part design through either adding material in critical areas of the part, reducing end-use loads or forgoing ME as a manufacturing option. To address this limitation, the authors propose a multi-axis deposition technique that deposits material along a part's surface to improve mechanical performance. Specifically, the authors employ a custom 6 degree of freedom robotic arm ME system to create a surface reinforcing ‘skin’, similar to composite layup, in a single manufacturing process. In this paper, vertical tensile bars are fabricated through stacked XY layers, followed by depositing material directly onto the printed surface to evaluate the effect of the skinning approach on mechanical properties. Experimental results demonstrate that surface-reinforced interlayer bonds provide increased yield strength.

The design and organization of complex robotic systems traditionally requires laborious trial-and-error processes to ensure both hardware and software components are correctly connected with the resources necessary for computation. This paper presents a novel generalization of the quadratic assignment and routing problem, introducing formalisms for selecting components and interconnections to synthesize a complete system capable of providing some user-defined functionality. By introducing mission context, functional requirements, and modularity directly into the assignment problem, we derive a solution where components are automatically selected and then organized into an optimal hardware and software interconnection structure, all while respecting restrictions on component viability and required functionality. The ability to generate \emph{complete} functional systems directly from individual components reduces manual design effort by allowing for a guided exploration of the design space. Additionally, our formulation increases resiliency by quantifying resource margins and enabling adaptation of system structure in response to changing environments, hardware or software failure. The proposed formulation is cast as an integer linear program which is provably $\mathcal{NP}$-hard. Two case studies are developed and analyzed to highlight the expressiveness and complexity of problems that can be addressed by this approach: the first explores the iterative development of a ground-based search-and-rescue robot in a variety of mission contexts, while the second explores the large-scale, complex design of a humanoid disaster robot for the DARPA Robotics Challenge. Numerical simulations quantify real world performance and demonstrate tractable time complexity for the scale of problems encountered in many modern robotic systems.

Mechatronics is a rapidly evolving technology that links mechanics, electronics and computer science to realize complex systems that improves our lives. The focal point for these systems is the microcontroller that links the digital/analog world to the mechanical. Inputs are provided by sensors and the output is typically actuation in some form. An overview of how these tools combine to form mechatronic systems is discussed. Sensing tools, actuation devices, bus structures for communicating with the microcontroller are discussed. To complement the hardware, is a wide variety of software programs that enhances the design process by minimizing the need for extensive bread boarding and prototyping.