Bojan Jerbić

Honey-bees optimization algorithm applied to path planning problem

Autonomous systems assume intelligent behaviour with capabilities of dealing in complex and changing environments. Problem of path planning, which can be observed as an optimization problem, seems to be of high importance for arising of intelligent behaviour for different real-world problem domains. Swarm intelligence has gained increasingly high interest among the researchers from different areas, like, science, commerce and engineering over the last few years. It is particularly suitable to apply methods inspired by swarm intelligence to various optimization problems, especially if the space to be explored is large and complex. This article presents application of Honey-bees mating algorithm (HBO) to a non linear Diophantine equation benchmark problem and comparison with results of a genetic algorithm (GA) designed for the same purpose. In second part of the work, HBO algorithm is applied to solve a problem of guidance of mobile robot through the space with differently shaped and distributed obstacles. Fuzzy fitness function for selective evaluation of paths found by the algorithm is proposed. The performance of the algorithm is comparable to genetic algorithm developed for the same purpose.

Context-Aware System Applied in Industrial Assembly Environment

The objective of this paper is to present an ongoing development of a context-aware system used within industrial environments. The core of the system is so-called Cognitive Model for Robot Group Control. This model is based on well-known concepts of Ubiquitous Computing, and is used to control robot behaviours in specially designed industrial environments. By using sensors integrated within the environment, the system is able to track and analyse changes, and update its informational buffer appropriately. Based on freshly collected information, the Model is able to provide a transformation of high-level contextual information to lower-level information that is much more suitable and understandable for technical systems. The Model uses semantically defined knowledge to define domain of interest, and Bayesian Network reasoning to deal with the uncertain events and ambiguity scenarios that characterize our naturally unstructured world.

Robotic application in neurosurgery using intelligent visual and haptic interaction

Today, the complexity and high technical requirements of neurosurgical operations are so demanding that modern robotic achievements and advances of accompanied technologies appear as the immanent means, which can significantly improve neurosurgical practice. A novel robotic system (RONNA –RObotic NeuroNAvigation) for application in neurosurgery is presented. The RONNA consists of two conventional articulated robot arms with a total of 13 degrees of freedom. A rigid and accurate robot is used for precise targeting of planned operating points and a compliant robot is used as operative assistant. A distinctive marker was developed for the purpose of precise localization and registration of the patient’s head. A novel visual calibration method is presented. The developed dual arm neurosurgical system enables flexible and reliable application with embedded behaviour based control providing intuitive interaction with surgical team and new possibilities compared to the existing surgical robot solutions.

Coordination of robots with overlapping workspaces based on motion co-evolution

The level of autonomy is the most important feature by which the modern robotic systems development is directed. Furthermore, if the robots are supposed to work together in order to solve a complex task, their workspaces are shared. In this case, the robots present dynamic obstacle to each other. This paper presents a solution of the problem of motion coordination of two robots with overlapping workspaces based on co-evolutionary algorithm for simultaneous motion planning of the two robots. A method for exact calculation of the solution coding chromosome length based on physical limitations of the robots is proposed. The algorithm is evaluated in a simulation environment developed in Matlab. Implementation to the real real industrial FANUC Lr Mate 200iC robots is performed. The simulation and implementation show high potential in terms of convergence robustness and time.

Industrial robotic system with adaptive control

In this paper an adaptive multiagent robotic assembly system is presented. State of the art industrial equipment is utilized to perform various assembly tasks in a highly unstructured environment without the need for central control. The emphasis is given to the developed methods that address particular issues in such robotic assembly systems. Close collaboration and intertwined work with human operators is one application under development, possible due to complex sensorial inputs on the robots. Active voice commands and prompts additionally contribute to human-robot interaction. Encounter with unknown objects is another issue that has been addressed and can be solved autonomously for simple case scenarios. Actual assembly applications as well as applications under development are presented. The operation in unstructured environments has been facilitated with vision systems, F/T sensors and other sensorial devices.

Dual-arm robot motion planning based on cooperative coevolution

This paper presents a cooperative coevolutionary approach to path planning for two robotic arms sharing common workspace. Each arm is considered an agent, required to find transition strategy from given initial to final configuration in the work space. Since the robots share workspace, they present dynamic obstacle to each other. To solve the problem of path planning in optimized fashion, we formulated it to multi-objective optimization domain and implemented co-evolutionary algorithm to simultaneously optimize four conflicting objectives. End-effector trajectory length, end-effector velocity distribution, total rotate angle and number of collisions are the objectives to be optimized. Simulation results for two 2-R type robots are presented.

A multiagent framework for industrial robotic applications

Abstract The paper presents a novel approach toward modeling and governing complex system behavior in flexible and adaptive robotic assembly systems. A fully distributed multiagent approach is implemented for autonomous control. The system is defined at multiple levels of granularity where agents provide services in respect to the current global goal. A decentralized multiagent approach is adopted for reasons of flexibility and fault tolerance embedded in the design phase. To prove the concept a robotic application for intelligent assembly is presented and discussed. It consists of multiple industrial robots equipped with force/torque sensors, 2D and 3D vision systems, automatic tool changers and other sensors and actuators. Through fusion of sensory input and mutual communication agents construct and negotiate an assembly plan and reconfigure respectively.

Swarm-Based Approach to Path Planning using Honey-Bees Mating Algorithm and ART Neural Network

In this paper, an integration of Honey bees mating algorithm (HBMA) and adaptive resonance theory neural network (ART1) for efficient path planning of a mobile robot in a static environment is presented. The robot must find shortest route from given origin to the target position. Moreover, it should be able to memorize the environment and, if it faces known world, execute already learned trajectory found by HBMA solver, or solve the world and memorize the trajectory for the given environment. This is done using Adaptive Resonance Theory based neural network. This way simulated robot is able to navigate through environment and to continuously increase its knowledge.

Medical applicability of a low-cost industrial robot arm guided with an optical tracking system

Robot systems used in surgical procedures can autonomously position tools at points correlated with preoperative imaging techniques such as magnetic resonance (MR) and computed tomography (CT). The aim of this paper is to measure and assess medical applicability of a low-cost, lightweight industrial robot arm (Universal robot UR5) guided with the medically certified optical tracking system (Polaris Vicra) to positions registered from a CT scan. Technical setup, measurement equipment, device communication and robot control based on OTS feedback are described. Robot intrinsic accuracy, CT scan accuracy and two methods of robot tool positioning with aid of the optical tracking system (OTS) are measured. Measurements show RMS error of the robot (0.669 mm) is decreased 55.4% when guided with OTS using a single marker probe (0.29 mm) and 40.5% when using OTS with relative referencing (0.39 mm). RMS error of the CT scan readings is 0.46 mm.

Autonomous robotic assembly using collaborative behavior based agents

This work studies the design of a robotic assembly system as a multiagent system. Any multidevice system, or any system whose performance is naturally decomposable, can be interpreted as a corporation of agents. Such a scheme comprises the ability of creating a collaborative technical system which can provide the achieving of a social intelligence. Social behavior is the highest form of intelligence that can provide the solving of very complex problems, autonomous creation of new procedures, and efficient adaptation to new tasks. The presented multiagent model is based on processing units, which include the recognition networks, problem solving algorithms, and learning engines. It integrates perception, recognition, learning, and communication capabilities. The reinforcement learning method is used here to evaluate robot behavior and to induce new, or improve existing, knowledge. The acquired action (task) plan is stored as experience which can be used in solving similar future problems. To provide the recognition of problem similarities, the Adaptive Fuzzy Shadowed (AFS) neural network is applied.