Small mobile robots equipped with long, extendable booms can perform tasks that require a large workspace—such as cable stringing and assembly for lunar construction.
Future infrastructure construction on the lunar surface will require semi- or fully-autonomous operation from robots deployed at the build site. In particular, tasks such as electrical outfitting necessitate transport, routing, and fine manipulation of cables across large structures. To address this need, we present a compact and long-reach manipulator incorporating a deployable composite boom, capable of performing manipulation tasks across large structures and workspaces. We characterize the deflection, vibration, and blossoming characteristics inherent to the deployable structure, and present a manipulation control strategy to mitigate these effects. Experiments indicate an average endpoint accuracy error of less than 15 mm for boom lengths up to 1.8 m. We demonstrate the approach with a cable routing task to illustrate the potential for lunar outfitting applications that benefit from long reach.
A rollable fiberglass boom (from Metolius Climbing) is integrated in a custom long-reach manipulator design. key hardware components include a shoulder base, deployable boom arm (0.2 - 3m), and dexterous wrist. Mechanical characterizations of the boom are performed for uniaxial bending, free vibration effects, and deployment blossoming effects.
The overall manipulator is modeled using a 6 DoF rigid kinematic model (RRPRRR). However, an additional transformation is used to augment the links after the prismatic (boom) joint in order to compensate for elastic boom deflection.
We develop a model-based control framework for closed-loop endpoint position control. The robot follows task trajectories using visual end-effector localization (from endpoint camera) and a task-space kinematic model (augmented with Euler beam deflections).
We consider the tracking errors across the robot workspace for a 10 cm × 10 cm square reference trajectory. The robot's executed trajectories across trials are overlaid, and the maximum spatial error is visualized as a task tube. Errors increase with both boom pitch angle and extension, especially at higher speeds.
The empirical model is applied to select appropriate task speeds for an autonomous cable routing task. At each of the four anchor locations, we select the appropriate task speed to maintain a position error under 15 mm. This allows for successful cable routing in different locations and orientations throughout the robot workspace.
