The ability to navigate complex unstructured environments and carry out inspection tasks requires robots to be capable of climbing inclined surfaces and to be equipped with a sensor payload. These features are desirable for robots that are used to inspect and monitor offshore energy platforms. Existing climbing robots mostly use rigid actuators, and robots that use soft actuators are not fully untethered yet. Another major problem with current climbing robots is that they are not built in a modular fashion, which makes it harder to adapt the system to new tasks, to repair the system, and to replace and reconfigure modules. This work presents a 450 g and a 250 • 250 • 140 mm modular, untethered hybrid hard/soft robot-Limpet II. The Limpet II uses a hybrid electromagnetic module as its core module to allow adhesion and locomotion capabilities. The adhesion capability is based on negative pressure adhesion utilizing suction cups. The locomotion capability is based on slip-stick locomotion. The Limpet II also has a sensor payload with nine different sensing modalities, which can be used to inspect and monitor offshore structures and the conditions surrounding them. Since the Limpet II is designed as a modular system, the modules can be reconfigured to achieve multiple tasks. To demonstrate its potential for inspection of offshore platforms, we show that the Limpet II is capable of responding to different sensory inputs, repositioning itself within its environment, adhering to structures made of different materials, and climbing inclined surfaces.

Operations in extreme and hostile environments, such as offshore oil and gas production, nuclear decommissioning, nuclear facility maintenance, deep mining, space exploration, and subsea applications, require the execution of sophisticated tasks. In nuclear environments, robotic systems have advanced significantly over the past years but still suffer from task failures caused by informational and physical uncertainty of the highly unstructured nature of the environment and exasperated by the time constraints imposed by high radiation levels. Herein, a survey is presented of current robotic systems that can operate in such extreme environments and offer a novel approach to solving the challenges they impose, encapsulated by the mission statement of providing structure in unstructured environments and exemplified by a new self‐assembling modular robotic system, the Connect‐R.

Soft robots are a new class of systems being developed and studied by robotics scientists. These systems have a diverse range of applications including sub-sea manipulation and rehabilitative robotics. In their current state of development, the prevalent paradigm for the control architecture in these systems is a one-to-one mapping of controller outputs to actuators. In this work, we define functional blocks as the physical implementation of some discrete behaviors, which are presented as a decomposition of the behavior of the soft robot. We also use the term ‘stacking’ as the ability to combine functional blocks to create a system that is more complex and has greater capability than the sum of its parts. By stacking functional blocks a system designer can increase the range of behaviors and the overall capability of the system. As the community continues to increase the capabilities of soft systems—by stacking more and more functional blocks—we will encounter a practical limit with the number of parallelized control lines. In this paper, we review 20 soft systems reported in the literature and we observe this trend of one-to-one mapping of control outputs to functional blocks. We also observe that stacking functional blocks results in systems that are increasingly capable of a diverse range of complex motions and behaviors, leading ultimately to systems that are capable of performing useful tasks. The design heuristic that we observe is one of increased capability by stacking simple units—a classic engineering approach. As we move towards more capability in soft robotic systems, and begin to reach practical limits in control, we predict that we will require increased amounts of autonomy in the system. The field of soft robotics is in its infancy, and as we move towards realizing the potential of this technology, we will need to develop design tools and control paradigms that allow us to handle the complexity in these stacked, non-linear systems.

No abstract

We introduce vPlanSim, an open source tool to aid in AI PDDL development. This tool is primarily aimed at researchers and developers who need a visual representation of their planning problem so that they can make useful insights into the performance of their system, and also to naturally convey their system to others. It is an open-source tool which allows a user to quickly and easily visualise a target environment to generate the problem files and also to visualise a plan. It is particularly well suited to spatial planning problems. This paper will demonstrate vPlanSim on 2D and 3D planning problems. vPlanSim is based on a small and carefully considered set of dependencies such as VTK and PyQt. It can be set up on different platforms and compiled from source with minimal effort. The code is and maintained via a clear code review mechanism. We welcome contributions from the open-source community.

Peristaltic conveyance can be used for the sorting and transport of delicate and nonrigid objects such as meat or soft fruit. The non‐linearity and stochastic behavior of peristaltic systems make them difficult to control. Optimizing controllers using machine learning represents a promising path to effective peristaltic control but currently, there is no suitable simulated model of a peristaltic table in which to run these optimizations. A simple, simulated model of a peristaltic conveyor that can be used for optimizing peristaltic control on a variety of peristaltic tables is presented. This simulator is demonstrated through a limited control problem evaluated on our real‐world system that is built for peristaltic conveyance. This simulator is available as the python package PeriSim so that it can be used by the robotics community for peristaltic control development.

In this paper, we offer a novel AI planning representation, based on a Cartesian coordinate system, for enabling the autonomous operations of Multi-Robot Systems in 3D environments. Each robot in the system has to conform to unique actuation and connection constraints that create a complex set of valid configurations. Our approach allows Multi-Robot Systems to self-assemble themselves into larger structures via AI planning, with the overarching goal of providing structural capabilities in harsh and uncertain environments. In comparing four different PDDL (Planning Domain Definition Language) domain representations, we show that our novel formulation satisfies the practical requirements emerging from robot deployment in the real world, resulting in an AI planning system that is accurate and efficient. We scale up performance by implementing direct FDR (Finite Domain Representation) generation based on the best performing PDDL model, bypassing the PDDL-to-FDR translation used by the majority of modern planners. The proposed approach is general and can be applied to a broad range of AI problems involving reasoning in 3D spaces.