Scenarios of Swarm Robotics

Hamann, Heiko

doi:10.1007/978-3-319-74528-2_4

Cited by 9 publications

(9 citation statements)

References 133 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since swarm robotic systems lack a centralized unit to control the entire swarm, careful design of individual controllers becomes vital in achieving the desired swarm behaviors. 23 Currently, the prevailing approach for designing the behavior of individual robots is the manual design method. 21 This method involves heuristically deducing individual behavior rules based on the swarm's target behavior and iteratively tuning an optimal controller.…”

Section: Related Workmentioning

confidence: 99%

An integrated Design Methodology for Swarm Robotics using Model-Based Systems Engineering and Robot Operating System

Aloui,

Guizani,

Hammadi

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

In swarm robotics, robots solve problems using collective behaviors similar to those observed in natural systems, such as birds, bees or fish. They determine their collective behavior through several simple interactions. However, the behavior of swarms emerging from local interactions remains difficult to predict. When trying to design swarm robotic systems for real applications, researchers are confronted with a wide range of software and hardware challenges in the different phases of the design process; problems related to the modeling of swarm systems, related to simulation, related to implementation on software and even on real systems, etc. Despite the increasing popularity of swarm robotics, designing effective and scalable swarm systems remains a challenging task due to the complex interactions between the individual robots and the environment. Therefore, there is a need for a systematic and comprehensive methodology that can guide designers in developing swarm robotics systems that are efficient, reliable, and adaptable to different scenarios. Our main contribution is to propose a new integrated methodology for the development of swarm robot systems. This methodology is based on modeling with Model-Based Systems Engineering method (MBSE) to specify the requirements and the collective behaviors of the swarms, then on the verification of the developed models and finally on the validation of the swarm system by physical prototyping with real robots. Our contribution also focuses on the development of a new SysML profile using the Domain Specific Language (DSL) that we call SwarmML to customize the functional and structural modeling of a swarm system (properties and attributes of the swarm). Two case studies are applied to validate our methodology; a case study of an aggregation of a swarm of robots and a case study of a collaborative simultaneous localization and mapping application (C-SLAM) performed by a swarm of Turtlebots. The novelty of this proposed methodology is the combination of SysML and Robot Operating System (ROS) to address the management of traceability between the different levels of swarm system design, in order to achieve functional, physical and software integration.

show abstract

Section: Related Workmentioning

confidence: 99%

An integrated Design Methodology for Swarm Robotics using Model-Based Systems Engineering and Robot Operating System

Aloui,

Guizani,

Hammadi

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

show abstract

“…Figure 11. A major focus of the collective intelligence field is to study the group intelligence and behaviors emerged from a large collection of individuals, whether in humans (Tapscott and Williams, 2008), animals (Sumpter, 2010), insects (Dorigo et al 2000; Seeley, 2010), or artificial swarm robots (Hamann, 2018; Rubenstein et al 2014). This focus has clearly been missing in the Deep RL field.…”

Section: Collective Intelligence For Deep Learningmentioning

confidence: 99%

“…Although the focus of much of the work in Deep RL is in learning neural network policies for an agent with a fixed design (e.g., a bipedal robot, humanoid, or robot arm), embodied intelligence is an area that is gathering interest in the sub-field (Ha, 2018; Pathak et al 2019). Inspired by several previous works on self-configuring modular robots by (Stoy et al 2010; Rubenstein et al 2014; Hamann, 2018), (Pathak et al 2019) investigates a collection of primitive agents that learn to self-assemble into a complex body while also learning a local policy to control the body without an explicit centralized control unit. Each primitive agent (which consists of a limb and a motor) can link up with nearby agents, allowing for complex morphologies to emerge.…”

Section: Collective Intelligence For Deep Learningmentioning

confidence: 99%

Collective intelligence for deep learning: A survey of recent developments

Tang

2022

Collective Intelligence

View full text Add to dashboard Cite

In the past decade, we have witnessed the rise of deep learning to dominate the field of artificial intelligence. Advances in artificial neural networks alongside corresponding advances in hardware accelerators with large memory capacity, together with the availability of large datasets enabled practitioners to train and deploy sophisticated neural network models that achieve state-of-the-art performance on tasks across several fields spanning computer vision, natural language processing, and reinforcement learning. However, as these neural networks become bigger, more complex, and more widely used, fundamental problems with current deep learning models become more apparent. State-of-the-art deep learning models are known to suffer from issues that range from poor robustness, inability to adapt to novel task settings, to requiring rigid and inflexible configuration assumptions. Collective behavior, commonly observed in nature, tends to produce systems that are robust, adaptable, and have less rigid assumptions about the environment configuration. Collective intelligence, as a field, studies the group intelligence that emerges from the interactions of many individuals. Within this field, ideas such as self-organization, emergent behavior, swarm optimization, and cellular automata were developed to model and explain complex systems. It is therefore natural to see these ideas incorporated into newer deep learning methods. In this review, we will provide a historical context of neural network research’s involvement with complex systems, and highlight several active areas in modern deep learning research that incorporate the principles of collective intelligence to advance its capabilities. We hope this review can serve as a bridge between the complex systems and deep learning communities.

show abstract

“…How the formation of the smooth low-density exterior with its coarse-grain, high density interior within a meandering macroscopic structure can be modelled mathematically is another aspect of interest. Granular convection as observed in the Brazil nut effect and fluid flow convection (coffee-ring effect) [64][65][66] paired with multi-scale reaction and diffusion processes (for the macroscopic structure) [46,47] may provide a model to approximate the observed geometry. If so, while the physical processes in fluid and granular material are similar, their connection to termite building is not obvious but may relate to a higher-arching self-organization/ self-segregation principle connecting physics of solids and fluids with that of behaviour in eusocial insects.…”

Section: Fine Grain Termite Bricksmentioning

confidence: 99%

Submillimetre mechanistic designs of termite-built structures

Oberst

Martin

Halkon

et al. 2021

J. R. Soc. Interface.

View full text Add to dashboard Cite

Termites inhabit complex underground mounds of intricate stigmergic labyrinthine designs with multiple functions as nursery, food storage and refuge, while maintaining a homeostatic microclimate. Past research studied termite building activities rather than the actual material structure. Yet, prior to understanding how multi-functionality shaped termite building, a thorough grasp of submillimetre mechanistic architecture of mounds is required. Here, we identify for Nasutitermes exitiosus via granulometry and Fourier transform infrared spectroscopy analysis, preferential particle sizes related to coarse silts and unknown mixtures of organic/inorganic components. High-resolution micro-computed X-ray tomography and microindentation tests reveal wall patterns of filigree laminated layers and sub-millimetre porosity wrapped around a coarse-grained inner scaffold. The scaffold geometry, which is designed of a lignin-based composite and densely biocementitious stercoral mortar, resembles that of trabecula cancellous bones. Fractal dimension estimates indicate multi-scaled porosity, important for enhanced evaporative cooling and structural stability. The indentation moduli increase from the outer to the inner wall parts to values higher than those found in loose clays and which exceed locally the properties of anthropogenic cementitious materials. Termites engineer intricately layered biocementitious composites of high elasticity. The multiple-scales and porosity of the structure indicate a potential to pioneer bio-architected lightweight and high-strength materials.

show abstract

Scenarios of Swarm Robotics

Cited by 9 publications

References 133 publications

An integrated Design Methodology for Swarm Robotics using Model-Based Systems Engineering and Robot Operating System

An integrated Design Methodology for Swarm Robotics using Model-Based Systems Engineering and Robot Operating System

Collective intelligence for deep learning: A survey of recent developments

Submillimetre mechanistic designs of termite-built structures

Contact Info

Product

Resources

About