As the United States is on the cusp of returning astronauts to the Moon, it becomes increasingly apparent that the assembly of structures in space will have to rely upon robots to perform the construction process. With a focus on sustaining a presence on the Moon's surface in such a harsh and unforgiving environment, demonstrating the robustness of autonomous assembly and capabilities of robotic manipulators is necessary. Current robotic assembly on Earth consists mainly of inspection or highly controlled environments, and always with a human in the loop to step in and fix issues if a problem occurs. To remove the human element, the robot system then must account for safety as well. Thus, system risk can easily overwhelm project costs. This thesis proposes a combination of hardware and state estimation solutions to improve the feasibility of low-fidelity and low-resolution robots for precision assembly tasks. Doing so reduces the risk to mission success, as the hardware becomes easier to replace or repair. The hardware modifications implement an electro-permanent magnet interface with alignment features to reduce the fidelity needed for the robot end effector. On the state estimation side, an Ensemble Kalman Filter is implemented, along with a scaling system to prevent FASER Lab hardware from becoming stuck due to hardware limitations. Overall, the three modifications improved the test robot's autonomous convergence error by 98.5%, bettering the system sufficiently to make an autonomous assembly process feasible. / Master of Science / With the dawn of new space age nearly upon us, one of the most important aspects to working in space will be robotic assembly, whether on the surface of other planetary bodies like the Moon or in zero-gravity, in order to keep astronauts safe and to reduce spaceship launch costs. Both places have their own difficult problems to deal with, and doing any actions in those locations come with a significant amount of risk involved. To reduce extreme risk, you can spend more money to over-protect the robots, or reduce the consequences of the risk. This thesis describes a way to reduce the impact of risks to a mission by checking whether inexpensive robots can be adapted and modified to be able to perform similar construction actions to a much more expensive robot. It does this by using specialized hardware and software programs to better align the robot to where it needs to go without people needing to step in and help it. The experiments showed a 98.5% improvement to the system from without any of the modifications and validated that the low-cost robot could be improved sufficiently to be useful.

This thesis investigates the implementation of a passive isolator with a pressurized air cushion for spacecraft payloads in mission architectures implementing in-space assembly technologies. A pressurized air bed capable of briefly surviving the space environment for cargo delivery was prototyped and experimentally evaluated for launch vehicle vibration dynamics resulting in a 72%, 93%, and 88% reduction in experienced GRMS loads for the X-Axis, Y-Axis, and Z-Axis, respectively. A preliminary Total Mass Loss evaluation of the Low-Density Polyethylene Film utilized for the air bed resulted in a mass loss of 0.7%, indicating that commercial off-the-shelf films might require minimal modification for flight readiness. An analytical model of a planar rectangular payload experiencing free vibrations with a Winkler foundation is generated and compared to the experimental results, showing a potential way for characterizing and designing such a foundation to reduce experienced vibrations. These preliminary results show a potential path for a non-cost-prohibitive method for space payloads to reduce loads experienced during launch as inspired by the successful hosted payloads program aboard the International Space Station.

The culmination of this research focuses on the area of structural support and stability as it relates to the field of large space structures. Fitting into the branch of in-space assembly, servicing, and manufacturing (ISAM), this topic covers essential subject matter areas of robotic manipulation, repair, state estimation, and structural health. As the next generation of space structures includes increased areas of modularity, the nature of structures built in-space lends itself significantly to repair efforts. With plans for these repair efforts in place, the lifetime of damaged structures can be greatly extended leading to a greater chance of mission success. By considering how repair efforts factor into the assembly scope, critical failures in large trusses, especially those involving single-point structural failures, can be mitigated. To do this, external forces are applied to the damaged structure utilizing an intelligent scaffolding formulation. This methodology employs robots to strategically apply loads to re-route abnormal stress and strain paths, correct for resulting deflections, and stabilize the structure itself. These tasks are vital to the safety of the structure and must take place before any repair efforts are considered in an effort to prevent cascading damage. The following research explores this damage simulation and correction paradigm through a variety of truss initial conditions, which allow for a suite of deflection responses. Utilizing these deflection responses a safe path for applying loads incrementally through generated waypoints is created with the help of the finite element modeler Ansys and a Python script. The ability for this system to successfully realign the wide scope of truss cases showcases that it is a truly adaptive system. Although this work is primarily proven within a simulation space, efforts to contextualize in a physical system and explore the elements needed to implement this method are also described. Finally, although this research is presented within the scope of damage repair, the final chapter looks to apply this method to other similarly unsupported structures by examining how critical it can be during assembly scenarios. / Doctor of Philosophy / As the industry sits on the edge of new in-space assembly technologies, a need to maintain these systems has arisen. The backbone upon which these new space technologies exist is with truss frameworks. By being able to build these sparse structures, large structures can be made with few components. These structures serve as critical mounting support for various instruments, engines, communication devices, solar panels, and more. As these structures are so critical across the board being able to repair a member when it becomes damaged is crucial. This research provides an avenue to do this. When a structure becomes damaged it will start to deform and bend. This presents unique challenges in attempting to replace an element or return the structure back to an operational state. First, the structure must be driven back into alignment to prevent further damage and hold the truss steady in preparation for repair. The methodology laid out within the dissertation covers how the use of simulations and force solvers can be implemented to create a path that allows a robot to force a structure into a desired configuration. To mimic the original damaged strut a strong stable platform robot called a Stewart Platform applies specific loads to correct the structure. This research shows structures with different specifications to highlight how this system can be universally applied to a single member damage scenario. Within this, simulations to showcase how this forcing method can be applied to these varying structures produce unique correction paths. These correction paths must be accurately driven across to ensure the safety of the structure. To broaden the application, a use case for this active force implementation was also shown as a critical component for assembly steps when elements are not properly supported. Throughout this, the need for intelligent scaffolding is shown to be a critical step in addressing structural health.

As exploration in space increases through the use of larger telescopes, more sophisticated structures, and physical exploration, the use of autonomous robots will become instrumental to build and maintain the infrastructures required for this exploration. These systems must be autonomous to deal with the infeasibility of teleoperation due signal delay and task complexity. The reality of using robots in the real world without direct human input will require the autonomous systems to have the capability of responding to errors that occur in an assembly scenario on their own. As such, a system must be in place to allow for the sequencing and allocation of tasks to the robotic workforce autonomously, giving the ability to re-plan in real world stochastic environments. This work presents four contributions towards a system allowing for the autonomous sequencing and allocation of tasks for in-space assembly problems. The first contribution is the development of the Stochastic Assembly Problem Definition (SAPD) to articulate all of the features in an assembly problem that are applicable to the task sequencing and allocation. The second contribution is the formulation of a mixed integer program to solve for assembly schedules that are optimal or a quantifiable measurement from optimal. This contribution is expanded through the development of a genetic algorithm formulation to utilize the stochastic information present in the assembly problem. This formulation extends the state-of-the-art techniques in genetic algorithms to allow for the inclusion of new constraints required for the in-space assembly domain. The third contribution addresses how to estimate a robot's ability to complete a task if the robot must be assigned to a task it was previously not expected to work on. This is accomplished through the development of four metrics and analyzed through the use of screw theory kinematics. The final contribution focuses on a set of metrics to guide the selection of a good scheduling method for different assembly situations. The experiments in this work demonstrate how the developed theory can be utilized and shows the scheduling systems producing the best or close to the best schedules for assemblies. It also shows how the metrics used to quantify and estimate robot ability are applied. The theory developed in this work provides another step towards autonomous systems that are capable of assembling structures in-space without the need for human input. / Doctor of Philosophy / As space exploration continues to progress, autonomous robots are needed to allow for the necessary structures to be built in-space, on Mars, and on the Lunar surface. Since it is not possible to plan for every possible thing that could go wrong or break, the robots must be able to figure out how to build and repair structures without human input. The work presented here develops a framework that allows this in-space assembly problem to be framed in a way the robots can process. It then provides a method for generating assembly schedules that describe very good, if not the best way to complete the assembly quickly while still taking into account randomness that may be present. Additionally, this work develops a way to quantify and estimate how good robots will be at a task they have not attempted before. Finally, a set of considerations are proposed to aid in determining what scheduling method will work best for different assembly scenarios. The experiments in this work demonstrate how the developed theory can be used and shows the scheduling systems producing the best or close to the best schedules for assemblies. It also shows how the methods used to define robot ability are applied. The work developed here provides another step towards autonomous systems that are capable of assembling structures in-space without the need for human input.

This research focuses on the development of the semantic and fiducial aided graph simultaneous localization and mapping (SF-GraphSLAM) method that is tailored for robotic assembly and servicing of large truss structures. SF-GraphSLAM contributes to the state of the art by creating a novel way to add associations between map landmarks, in this scenario fiducials, by pre-generating a semantic map of expected relations based on the truss module known models, kinematic information about deployable modules, and hardware constraints for assembled modules. This additional information about the expected fiducial relations, and therefore truss module relative poses, can be used to add robustness to camera pose and measurement error. In parallel, the concept of a mixed assembly truss structure paradigm was created and analyzed. This mixed assembly method focuses on reducing the number of modules required to construct large truss structures by using a mixture of deployable and assembled truss modules to create a checkerboard array that is scalable to various dimensions and shapes while still minimizing the number of modules compared to a strut-by-strut method. Leveraging this paradigm SF-GraphSLAM is able to start at an advantage in terms of minimizing the state vector for the assembly testing. In addition, due to the added knowledge of the structure and the choice to utilize fiducial markers, SF-GraphSLAM is able to minimize the number of fiducials used to define the structure and therefore have the minimum state space to solve the assembly scenario, greatly improving the real-time estimation process between assembly steps. These optimizations will have a larger effect as the size of the scaled end structure increases. SF-GraphSLAM is derived in mathematical form following the same core process used for the pose and measurement components used in the base GraphSLAM. SF-GraphSLAM is evaluated against the state of the art example of GraphSLAM through simulation using an example 3x3x3 mixed assembly truss structure, known as the Built On-orbit Robotically-assembled Gigatruss (BORG). A physical BORG test truss was constructed to enable hardware trials of the SF-GraphSLAM algorithm. Although this ground hardware is not ideal for the high precision application of space structures it allows for rapid experimental robotic testing. This tailored SF-GraphSLAM approach will contribute to the state of the art of robotic in-space servicing, assembly, and manufacturing (ISAM) by providing progress on a method for dealing with the autonomous robotic assembly of movable modules to create larger structures. This will be critical for missions such as robotically assembling a large antenna structure or space telescope. Furthermore, the core methodology will study into how to best utilize information in a large-scale structure environment, including non-static flexible or deployable modules, to adequately map it which is also applicable to the larger field of robotic operations dealing with structures such as bridge surveying. / Doctor of Philosophy / The goal of this research is to enable in-space assembly of large truss structures by advancing the state of the art of how a robot can map the structure it is actively assembling. The concept of having a robot create a map of the landmarks, or in this case truss elements, it sees while keeping track of it's own movement is known as simultaneous localization and mapping (SLAM). This research focuses on the creation of a method called semantic and fiducial aided graph simultaneous localization and mapping (SF-GraphSLAM). The added semantic information is the model knowledge of the truss structure the robot is assembling, including what kind of modules are within and their desired relationships to each other. Fiducials are optical markers that can be used to provide identification, position, and orientation of what they are mounted to. Combining these concepts SF-GraphSLAM can use easily identifiable fiducials to mark components of the truss structure and use knowledge of how the truss structure should be assembled to help in estimating where the actual physical components are at different stages of the assembly process. This method is used to check if a truss module is assembled correctly after each step to ensure the final structure is within the requirements desired. This concept can be likened to when assembling a LEGO model, a person verifies they are using the correct brick for the next assembly step and then compared the state of the model with the reference photo before proceeding with the building. An incorrectly assembled module in an early step could result in a module down the line not being able to be properly placed or the final assembled structure not being within operational tolerances. This research shows how SF-GraphSLAM can be implemented for the application of assembling a truss structure out of both deployable and assembled modules. Mathematical analysis, simulations, and hardware testing were completed to compare this new method to the state of the art approach. SF-GraphSLAM is a critical step in the development required to make autonomous robotic assembly of larger structures in space feasible.