Nuno M. F. Ferreira

An efficient method for segmentation of images based on fractional calculus and natural selection

Image segmentation has been widely used in document image analysis for extraction of printed characters, map processing in order to find lines, legends, and characters, topological features extraction for extraction of geographical information, and quality inspection of materials where defective parts must be delineated among many other applications. In image analysis, the efficient segmentation of images into meaningful objects is important for classification and object recognition. This paper presents two novel methods for segmentation of images based on the Fractional-Order Darwinian Particle Swarm Optimization (FODPSO) and Darwinian Particle Swarm Optimization (DPSO) for determining the n-1 optimal n-level threshold on a given image. The efficiency of the proposed methods is compared with other well-known thresholding segmentation methods. Experimental results show that the proposed methods perform better than other methods when considering a number of different measures.

A novel multi-robot exploration approach based on Particle Swarm Optimization algorithms

This paper proposes two extensions of Particle Swarm Optimization (PSO) and Darwinian Particle Swarm Optimization (DPSO), respectively named as RPSO (Robotic PSO) and RDPSO (Robotic DPSO), so as to adapt these promising biological-inspired techniques to the domain of multi-robot systems, by taking into account obstacle avoidance. These novel algorithms are demonstrated for groups of simulated robots performing a distributed exploration task. The concepts of social exclusion and social inclusion are used in the RDPSO algorithm as a “punish-reward” mechanism enhancing the ability to escape from local optima. Experimental results obtained in a simulated environment show that biological and sociological inspiration can be useful to meet the challenges of robotic applications that can be described as optimization problems (e.g. search and rescue).

Benchmark of swarm robotics distributed techniques in a search task

This paper presents a survey on multi-robot search inspired by swarm intelligence by further classifying and discussing the theoretical advantages and disadvantages of the existing studies. Subsequently, the most attractive techniques are evaluated and compared by highlighting their most relevant features. This is motivated by the gradual growth of swarm robotics solutions in situations where conventional search cannot find a satisfactory solution. For instance, exhaustive multi-robot search techniques, such as sweeping the environment, allow for a better avoidance of local solutions but require too much time to find the optimal one. Moreover, such techniques tend to fail in finding targets within dynamic and unstructured environments. This paper presents experiments conducted to benchmark five state-of-the-art algorithms for cooperative exploration tasks. The simulated experimental results show the superiority of the previously presented Robotic Darwinian Particle Swarm Optimization (RDPSO), evidencing that sociobiological inspiration is useful to meet the challenges of robotic applications that can be described as optimization problems (e.g., search and rescue). Moreover, the RDPSO is further compared with the best performing algorithms within a population of 14 e-pucks. It is observed that the RDPSO algorithm converges to the optimal solution faster and more accurately than the other approaches without significantly increasing the computational demand, memory and communication complexity.

A fuzzified systematic adjustment of the robotic Darwinian PSO

The Darwinian Particle Swarm Optimization (DPSO) is an evolutionary algorithm that extends the Particle Swarm Optimization using natural selection to enhance the ability to escape from sub-optimal solutions. An extension of the DPSO to multi-robot applications has been recently proposed and denoted as Robotic Darwinian PSO (RDPSO), benefiting from the dynamical partitioning of the whole population of robots, hence decreasing the amount of required information exchange among robots. This paper further extends the previously proposed algorithm adapting the behavior of robots based on a set of context-based evaluation metrics. Those metrics are then used as inputs of a fuzzy system so as to systematically adjust the RDPSO parameters (i.e., outputs of the fuzzy system), thus improving its convergence rate, susceptibility to obstacles and communication constraints. The adapted RDPSO is evaluated in groups of physical robots, being further explored using larger populations of simulated mobile robots within a larger scenario.

Ensuring ad hoc connectivity in distributed search with Robotic Darwinian Particle Swarms

This paper presents an enforcing multi-hop network connectivity algorithm experimentally validated using a modified version of the Darwinian Particle Swarm Optimization (DPSO), denoted as RDPSO (Robotic DPSO) on groups of simulated robots performing a distributed exploration task. This work aims to overcome limitations of multi-robot systems (MRS) in difficult scenarios (e.g., search and rescue) concerning the need and the ability to actively maintain an available inter-robot communication channel, through the development of effective multi-robot cooperation without relying on a preexisting communication network. Although there is no linear relationship between the number of robots (i.e., nodes) and the maximum communication range, experimental results show that the decreased performance by the developed algorithm under communication constraints can be overcome by slightly increasing the number of robots as the maximum communication range is decreased.

Use of Darwinian Particle Swarm Optimization technique for the segmentation of Remote Sensing images

In this work, a novel method for segmentation of Remote Sensing (RS) images based on the Darwinian Particle Swarm Optimization (DPSO) for determining the n-1 optimal n-level threshold on a given image is proposed. The efficiency of the proposed method is compared with the Particle Swarm Optimization (PSO) based segmentation method. Results show that DPSO-based image segmentation performs better than PSO-based method in a number of different measures.

Introducing the fractional order robotic Darwinian PSO

The Darwinian Particle Swarm Optimization (DPSO) is an evolutionary algorithm that extends the Particle Swarm Optimization using natural selection to enhance the ability to escape from sub-optimal solutions. An extension of the DPSO to multi-robot applications has been recently proposed and denoted as Robotic Darwinian PSO (RDPSO), benefiting from the dynamical partitioning of the whole population of robots, hence decreasing the amount of required information exchange among robots. This paper further extends the previously proposed algorithm using fractional calculus concepts to control the convergence rate, while considering the robot dynamical characteristics. Moreover, to improve the convergence analysis of the RDPSO, an adjustment of the fractional coefficient based on mobile robot constraints is presented and experimentally assessed with 2 real platforms. Afterwards, this novel fractional-order RDPSO is evaluated in 12 physical robots being further explored using a larger population of 100 simulated ...

Darwinian swarm exploration under communication constraints: Initial deployment and fault-tolerance assessment

In most real multi-robot applications, such as search-and-rescue, cooperative robots have to move to complete their tasks while maintaining communication among themselves without the aid of a communication infrastructure. However, initially deploying and ensuring a mobile ad-hoc network in real and complex environments is an arduous task since the strength of the connection between two nodes (i.e., robots) can change rapidly in time or even disappear. An extension of the Particle Swarm Optimization to multi-robot applications has been previously proposed and denoted as Robotic Darwinian PSO (RDPSO). This paper contributes with a further extension of the RDPSO, thus integrating two research aspects: (i) an autonomous, realistic and fault-tolerant initial deployment strategy denoted as Extended Spiral of Theodorus (EST); and (ii) a fault-tolerant distributed search to prevent communication network splits. The exploring agents, denoted as scouts, are autonomously deployed using supporting agents, denoted as rangers. Experimental results with 15 physical scouts and 3 physical rangers show that the algorithm converges to the optimal solution faster and more accurately using the EST approach over the random deployment strategy. Also, a more fault-tolerant strategy clearly influences the time needed to converge to the final solution, but is less susceptible to robot failures.

Mechanism and Convergence Analysis of a Multi-Robot Swarm Approach Based on Natural Selection

The Darwinian Particle Swarm Optimization (DPSO) is an evolutionary algorithm that extends the Particle Swarm Optimization (PSO) using natural selection, or survival-of-the-fittest, to enhance the ability to escape from local optima. An extension of the DPSO to multi-robot applications has been recently proposed and denoted as Robotic Darwinian PSO (RDPSO), benefiting from the dynamical partitioning of the whole population of robots. Therefore, the RDPSO decreases the amount of required information exchange among robots, and is scalable to large populations of robots. This paper presents a stability analysis of the RDPSO to better understand the relationship between the algorithm parameters and the robot’s convergence. Moreover, the analysis of the RDPSO is further extended for real robot constraints (e.g., robot dynamics, obstacles and communication constraints) and experimental assessment with physical robots. The optimal parameters are evaluated in groups of physical robots and a larger population of simulated mobile robots for different target distributions within larger scenarios. Experimental results show that robots are able to converge regardless of the RDPSO parameters within the defined attraction domain. However, a more conservative parametrization presents a significant influence on the convergence time. To further evaluate the herein proposed approach, the RDPSO is further compared with four state-of-the-art swarm robotic alternatives under simulation. It is observed that the RDPSO algorithm provably converges to the optimal solution faster and more accurately than the other approaches.

Marsupial teams of robots: deployment of miniature robots for swarm exploration under communication constraints

Mobile Ad hoc Networks have attracted much attention in the last years, since they allow the coordination and cooperation between agents belonging to a multi-robot system. However, initially deploying autonomously a wireless sensor robot network in a real environment has not taken the proper attention. Moreover, maintaining the connectivity between agents in real and complex environments is an arduous task since the strength of the connection between two nodes (i.e., robots) can change rapidly in time or even disappear. This paper compares two autonomous and realistic marsupial strategies for initial deployment in unknown scenarios, in the context of swarm exploration: Random and Extended Spiral of Theodorus . These are based on a hierarchical approach, in which exploring agents, named scouts , are autonomously deployed through explicit cooperation with supporting agents, denoted as rangers . Experimental results with a team of heterogeneous robots are conducted using both real and virtual robots. Results show the effectiveness of the methods, using a performance metric based on dispersion. Conclusions drawn in this work pave the way for a whole series of possible new approaches.