Nikolaus Correll

Social Integration of Robots into Groups of Cockroaches to Control Self-Organized Choices

Collective behavior based on self-organization has been shown in group-living animals from insects to vertebrates. These findings have stimulated engineers to investigate approaches for the coordination of autonomous multirobot systems based on self-organization. In this experimental study, we show collective decision-making by mixed groups of cockroaches and socially integrated autonomous robots, leading to shared shelter selection. Individuals, natural or artificial, are perceived as equivalent, and the collective decision emerges from nonlinear feedbacks based on local interactions. Even when in the minority, robots can modulate the collective decision-making process and produce a global pattern not observed in their absence. These results demonstrate the possibility of using intelligent autonomous devices to study and control self-organized behavioral patterns in group-living animals.

Materials that couple sensing, actuation, computation, and communication

Adding autonomy to materials science Shape-memory alloys can alter their shape in response to a change in temperature. This can be thought of as a simple autonomous response, albeit one that is fully programmed at the time of fabrication. It is now possible to build materials or combinations of materials that can sense and respond to their local environment, in ways that might also include simple computations and communication. McEvoy and Correll review recent developments in the creation of autonomous materials. They look at how individual abilities are added to a material and the current limitations in the further development of “robotic materials.” Science, this issue 10.1126/science.1261689 BACKGROUND The tight integration of sensing, actuation, and computation that biological systems exhibit to achieve shape and appearance changes (like the cuttlefish and birds in flight), adaptive load support (like the banyan tree), or tactile sensing at very high dynamic range (such as the human skin) has long served as inspiration for engineered systems. Artificial materials with such capabilities could enable airplane wings and vehicles with the ability to adapt their aerodynamic profile or camouflage in the environment, bridges and other civil structures that could detect and repair damages, or robotic skin and prosthetics with the ability to sense touch and subtle textures. The vision for such materials has been articulated repeatedly in science and fiction (“programmable matter”) and periodically has undergone a renaissance with the advent of new enabling technology such as fast digital electronics in the 1970s and microelectromechanical systems in the 1990s. ADVANCES Recent advances in manufacturing, combined with the miniaturization of electronics that has culminated in providing the power of a desktop computer of the 1990s on the head of a pin, is enabling a new class of “robotic” materials that transcend classical composite materials in functionality. Whereas state-of-the-art composites are increasingly integrating sensors and actuators at high densities, the availability of cheap and small microprocessors will allow these materials to function autonomously. Yet, this vision requires the tight integration of material science, computer science, and other related disciplines to make fundamental advances in distributed algorithms and manufacturing processes. Advances are currently being made in individual disciplines rather than system integration, which has become increasingly possible in recent years. For example, the composite materials community has made tremendous advances in composites that integrate sensing for nondestructive evaluation, and actuation (for example, for shape-changing airfoils), as well as their manufacturing. At the same time, computer science has created an entire field concerned with distributed algorithms to collect, process, and act upon vast collections of information in the field of sensor networks. Similarly, manufacturing has been revolutionized by advances in three-dimensional (3D) printing, as well as entirely new methods for creating complex structures from unfolding or stretching of patterned 2D composites. Finally, robotics and controls have made advances in controlling robots with multiple actuators, continuum dynamics, and large numbers of distributed sensors. Only a few systems have taken advantage of these advances, however, to create materials that tightly integrate sensing, actuation, computation, and communication in a way that allows them to be mass-produced cheaply and easily. OUTLOOK Robotic materials can enable smart composites that autonomously change their shape, stiffness, or physical appearance in a fully programmable way, extending the functionality of classical “smart materials.” If mass-produced economically and available as a commodity, robotic materials have the potential to add unprecedented functionality to everyday objects and surfaces, enabling a vast array of applications ranging from more efficient aircraft and vehicles, to sensorial robotics and prosthetics, to everyday objects like clothing and furniture. Realizing this vision requires not only a new level of interdisciplinary collaboration between the engineering disciplines and the sciences, but also a new model of interdisciplinary education that captures both the disciplinary breadth of robotic materials and the depth of individual disciplines. (Top) Biological systems that tightly integrate sensing, actuation, computation, and communication and (bottom) the engineering applications that could be enabled by materials that take advantage of similar principles. (From left) The cuttlefish (camouflage), an eagle’s wings (shape change), the banyan tree (adaptive load support), and human skin (tactile sensing). CREDITS: CUTTLEFISH: N. HOBGOOD/WIKIMEDIA COMMONS; BALD EAGLE ALASKA: C. CHAPMAN/WIKIMEDIA COMMONS; BANYAN TREE: W. KNIGHT/WIKIMEDIA COMMONS; HUMAN SKIN: A. MCEVOY; MEN IN CAMOUFLAGE HUNTING GEAR: H. RYAN/U.S. FISH AND WILDLIFE SERVICE; 21ST CENTURY AEROSPACE VEHICLE: NASA; SYDNEY HARBOUR BRIDGE: I. BROWN/WIKIMEDIA COMMONS; CYBERHAND: PRENSILIA S.R.L/ PRENSILIA.COM Tightly integrating sensing, actuation, and computation into composites could enable a new generation of truly smart material systems that can change their appearance and shape autonomously. Applications for such materials include airfoils that change their aerodynamic profile, vehicles with camouflage abilities, bridges that detect and repair damage, or robotic skins and prosthetics with a realistic sense of touch. Although integrating sensors and actuators into composites is becoming increasingly common, the opportunities afforded by embedded computation have only been marginally explored. Here, the key challenge is the gap between the continuous physics of materials and the discrete mathematics of computation. Bridging this gap requires a fundamental understanding of the constituents of such robotic materials and the distributed algorithms and controls that make these structures smart.

SwisTrack - a flexible open source tracking software for multi-agent systems

Vision-based tracking is used in nearly all robotic laboratories for monitoring and extracting of agent positions, orientations, and trajectories. However, there is currently no accepted standard software solution available, so many research groups resort to developing and using their own custom software. In this paper, we present version 4 of SwisTrack, an open source project for simultaneous tracking of multiple agents. While its broad range of pre-implemented algorithmic components allows it to be used in a variety of experimental applications, its novelty stands in its highly modular architecture. Advanced users can therefore also implement additional customized modules which extend the functionality of the existing components within the provided interface. This paper introduces SwisTrack and shows experiments with both marked and marker-less agents.

Collective Inspection of Regular Structures using a Swarm of Miniature Robots

We present a series of experiments concerned with the inspection of regular, engineered structures carried out using swarms of five to twenty autonomous, miniature robots, solely endowed with onboard, local sensors. Individual robot controllers are behaviorbased and the swarm coordination relies on a fully distributed control algorithm. The resulting collective behavior emerges from a combination of simple robot-robot interactions and the underlying environmental template. To estimate intrinsic advantages and limitations of the proposed control solution, we capture its characteristics at higher abstraction levels using nonspatial, microscopic and macroscopic probabilistic models. Although both types of models achieve only qualitatively correct predictions, they help us to shed light on the influence of the environmental template and control design choices on the considered nonspatial swarm metrics (inspection time and redundancy). Modeling results suggest that additional geometric details of the environmental structure should be taken into account for improving prediction accuracy and that the proposed control solution can be further optimized without changing its underlying architecture.

Soft Autonomous Materials—Using Active Elasticity and Embedded Distributed Computation

The impressive agility of living systems seems to stem from modular sensing, actuation and communication capabilities, as well as intelligence embedded in the mechanics in the form of active compliance. As a step towards bridging the gap between man-made machines and their biological counterparts, we developed a class of soft mechanisms that can undergo shape change and locomotion under pneumatic actuation. Sensing, computation, communication and actuation are embedded in the material leading to an amorphous, soft material. Soft mechanisms are harder to control than stiff mechanisms as their kinematics are difficult to model and their degrees of freedom are large. Here we show instances of such mechanisms made from identical cellular elements and demonstrate shape changing, and autonomous, sensor-based locomotion using distributed control. We show that the flexible system is accurately modeled by an equivalent spring-mass model and that shape change of each element is linear with applied pressure. We also derive a distributed feedback control law that lets a belt-shaped robot made of flexible elements locomote and climb up inclinations. These mechanisms and algorithms may provide a basis for creating a new generation of biomimetic soft robots that can negotiate openings and manipulate objects with an unprecedented level of compliance and robustness.

SwisTrack: A Tracking Tool for Multi-Unit Robotic and Biological Systems

Tracking of miniature robotic platforms involves major challenges in image recognition and data association. We present our 2.5 years effort into developing a platform-independent, easy to use, and robust tracking software SwisTrack, which is tailored to research in swarm robotics and behavioral biology. We demonstrate the software and algorithms abilities using two case studies, tracking of a swarm of cockroaches, and a swarm-robotic inspection task, while outlining hard problems in tracking and data-association of marker-less objects. Its open, platform-independent architecture, and easy-to-use interfaces (Matlab, Java, and C++), allowing for (distributed) post-processing of trajectory data online, make the software highly adaptive to particular research projects without changes to the source code. SwisTrack will be publicly available shortly under the OSI Adaptive License via SourceForge.net.

Collaborative coverage using a swarm of networked miniature robots

We study distributed coverage of environments with unknown extension using a team of networked miniature robots analytically and experimentally. Algorithms are analyzed by incrementally raising the abstraction level starting from physical robots, to realistic and discrete event system (DES) simulation. The realistic simulation is calibrated using sensor and actuator noise characteristics of the real platform and serves for calibration of the DES microscopic model. The proposed algorithm is robust to positional noise and communication loss, and its performance gracefully degrades for communication and localization failures to a lower bound, which is given by the performance of a non-coordinated, randomized solution. Results are validated by real robot experiments with miniature robots of a size smaller than 2 cmx2 cmx3 cm in a boundary coverage case study. Trade-offs between the abilities of the individual platform, required communication, and algorithmic performance are discussed.

Multirobot inspection of industrial machinery

Inspection of aircraft and power generation machinery using a swarm of miniature robots is a promising application both from an intellectual and a commercial perspective. Our research is motivated by a case study concerned with the inspection of a jet turbine engine by a swarm of miniature robots. This article summarizes our efforts that include multirobot path planning, modeling of self-organized robotic systems, and the implementation of proof-of-concept experiments with real miniature robots. Although other research tackles challenges that arise from moving within three-dimensional (3-D) structured environments at the level of the individual robotic node, the emphasis of our work is on explicitly incorporating the potential limitations of the individual robotic platform in terms of sensor and actuator noise into the modeling and design process of collaborative inspection systems. We highlight difficulties and further challenges on the (lengthy) path toward truly autonomous parallel robotic inspection of complex engineered structures.

Robust Distributed Coverage using a Swarm of Miniature Robots

For the multi-robot coverage problem deterministic deliberative as well as probabilistic approaches have been proposed. Whereas deterministic approaches usually provide provable completeness and promise good performance under perfect conditions, probabilistic approaches are more robust to sensor and actuator noise, but completion cannot be guaranteed and performance is sub-optimal in terms of time to completion. In reality, however, almost all deterministic algorithms for robot coordination can be considered probabilistic when considering the unpredictability of real world factors. This paper investigates experimentally and analytically how probabilistic and deterministic algorithms can be combined for maintaining the robustness of probabilistic approaches, and explicitly model the reliability of a robotic platform. Using realistic simulation and data from real robot experiments, we study system performance of a swarm-robotic inspection system at different levels of noise (wheel-slip). The prediction error of a purely deterministic model increases when the assumption of perfect sensors and actuators is violated, whereas a combination of probabilistic and deterministic models provides a better match with experimental data.

Modeling and designing self-organized aggregation in a swarm of miniature robots

We model the dynamics of self-organized robot aggregation inspired by a study on the aggregation of gregarious arthropods. In swarms of German cockroaches, aggregation into clusters emerges solely from local interactions between the individuals, whereas the probabilities of joining or leaving a cluster are a function of the cluster size. We propose a non-spatial population dynamics model that keeps track of the number of robots in clusters of specific size using control parameters of the individual robots and the probability of detecting another robot in the environment. The model is able to quantitatively and qualitatively predict the dynamics observed in extensive realistic multi-robot simulation, and provides qualitative agreement with data obtained from aggregation of Blattela germanica larvae. In particular, we show by analysis, numerical and realistic simulation that the emergence of a single aggregate requires a minimal communication range between individuals.