Perception-Informed Autonomous Environment Augmentation with Modular Robots

Abstract

We present a system enabling a modular robot to autonomously build structures in order to accomplish highlevel tasks. Building structures allows the robot to surmount large obstacles, expanding the set of tasks it can perform. This addresses a common weakness of modular robot systems, which often struggle to traverse large obstacles.This paper presents the hardware, perception, and planning tools that comprise our system. An environment characterization algorithm identifies features in the environment that can be augmented to create a path between two disconnected regions of the environment. Specially-designed building blocks enable the robot to create structures that can augment the environment to make obstacles traversable. A high-level planner reasons about the task, robot locomotion capabilities, and environment to decide if and where to augment the environment in order to perform the desired task. We validate our system in hardware experiments.

Roombots extended: Challenges in the next generation of self-reconfigurable modular robots and their application in adaptive and assistive furniture

“…With Roombots, the focus is not only task execution with prepared structures, but also the actual formation process in 3D. While some of the systems in [2] and [21] [33] and [34]). We aim to show that Roombots is one of the few systems where 3D reconfiguration with 10 modules (30 DoF) and more is possible.…”

“…integrated in a module, a specialized module, an external system), each with their own advantages and disadvantages and engineering difficulties. A recent use of RGB-D sensor in the scope of SRMR has been reported in [33] where a system made out of SMORES carries the RGB-D sensor around and uses it for closed-loop self-reconfiguration as well as navigation. The simplest way for tracking however is to keep the RGB-D camera external.…”

A review of collective robotic construction

“…Alternatives include locally informed robots that either coordinate directly (58,59), broadcast to nearby neighbors (22), or are agnostic to other robots and use stigmergic coordination (11,14). Such systems have been demonstrated with ground-locked and climbing robots using light-based sensors and environmental templates (21,22), smart materials (25), passive mechanical and visual patterns on the material (14,64), depth scanners (11), and more traditional image processing techniques (24,65). As more advanced sensors such as radar, depth cameras, laser imaging and ranging systems, and GPS become cheaper and more readily accessible, they may play a bigger role in the field.…”

Modular Reconfigurable Robotics

“…Broadcast messages are particularly useful as they are not limited to an address space. Initial studies have used distributed sensing of the environment using bump switches (20) as well as mapping using onboard sensors (63), while combining with an external sensing system seems to be effective for achieving dedicated tasks, either through mobile (64) or global systems (5). In heterogeneous systems, some of the sensing tasks can be off-loaded to dedicated sensing modules (65,64).…”