We present the design, integration, and evaluation of a full-stack robotic system called RoMan, which can conduct autonomous field operations involving physical interaction with its environment. RoMan offers autonomous behaviors that can be triggered from succinct, high-level human input such as “open this box and retrieve the bag inside.” The robot’s behaviors are driven by a set of planners and controllers grounded in perceptual reconstructions of the environment. These behaviors are articulated by a behavior tree that translates high-level operator input into programs of increasing sensorimotor expressiveness, ultimately driving the lowest-level controllers. The software system is implemented in ROS as a set of independent processes connected by synchronous and asynchronous communication, and distributed across two on-board planning/control computers. The behavior stack drives a novel platform consisting of a pair of custom, 500 Nm/axis manipulators mounted on a rotatable torso aboard a tracked platform. The robot’s head is equipped with forward-looking depth cameras, and the arms carry wrist-mounted force-torque sensors and a mix of three- and four-finger grippers. We discuss design and implementation trade-offs affecting the entire hardware-software stack and high-level manipulation behaviors. We also demonstrate the applicability of the system for solving two manipulation tasks: 1) removing heavy debris from a roadway, where 64% of end-to-end autonomous runs required at most one human intervention; and 2) retrieving an item from a closed container, with a fully autonomous success rate of 56%. Finally, we indicate lessons learned and suggest outstanding research problems.
Continued advancements in robot autonomy have allowed the research community to shift from using robots as tools in the field to deploying robot teammates capable of learning, reasoning, and executing tasks. Autonomous navigation is one necessary capability of a robot teammate that must operate in large field environments. In relatively static environments a simple navigation solution such as obstacle avoidance along the shortest path may suffice; however, as robot teammates are deployed to highly dynamic environments with changing mission requirements, additional environment context may be necessary to ensure safe and reliable navigation. Although recent works in urban autonomous driving have advanced the state-of-the-art in context-aware decision making, the spectrum of behaviors deployed for context-switching is more narrowly focused (by defining constraints specific to operation in structured environments) than what might be required for human-agent teaming field missions. As such, establishing a context-aware intelligent system for dynamic, unstructured environments is still an open problem. We discuss our approach to the integration of several context-aware navigation behaviors on a small unmanned ground vehicle (UGV) and a perception stack that provides cues used to transition between these different learned behaviors. Specifically, we integrate socially compliant, terrain-aware, and covert behaviors in an outdoor navigation scenario where the UGV encounters moving pedestrians, different terrains, and weapon threats. We provide a detailed account of the overall system integration, experiment design, component- and system-level analysis, and lessons learned.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.