Mobile Robotic Agents Research Articles

A systematic strategy for the architecture design of collaborative and reconfigurable assembly lines

Current manufacturing environments have to face a competitive and evolving framework characterised by high-mix and low-volume demands. To properly deal with new requests coming from the market, Reconfigurable Assembly Systems (RASs) have to be designed. In this paper, we propose a methodology for the design of the architecture of a RAS, featuring both human and mobile robotic agents, and dealing with multi-product assembly of medium-light-weight products, originally assembled manually. In particular, a systematic methodology is proposed to define the number of workstations, specifying which set of operations must be performed in each one, and to design the system layout guaranteeing reconfigurability. To this end, the concepts of operation, datasheet of an operation, assembly tree, and macro-operation are introduced and formalised. Guidelines to automatise the required operations in a flexible way are also proposed. The method is applied to a use case where we define the proper number of workstations of a RAS to assemble several types of motorbike brakes, showing the reconfigurability of the system.

Distributed Formation Control of Multi-Robot Systems with Path Navigation via Complex Laplacian.

This paper focuses on the formation control of multi-robot systems with leader-follower network structure in directed topology to guide a system composed of multiple mobile robot agents to achieve global path navigation with a desired formation. A distributed linear formation control strategy based on the complex Laplacian matrix is employed, which enables the robot agents to converge into a similar formation of the desired formation, and the size and orientation of the formation are determined by the positions of two leaders. Additionally, in order to ensure that all robot agents in the formation move at a common velocity, the distributed control approach also includes a velocity consensus component. Based on the realization of similar formation control of a multi-robot system, the path navigation algorithm is combined with it to realize the global navigation of the system as a whole. Furthermore, a controller enabling the scalability of the formation size is introduced to enhance the overall maneuverability of the system in specific scenarios like narrow corridors. The simulation results demonstrate the feasibility of the proposed approach.

Ferrobotic swarms enable accessible and adaptable automated viral testing.

Expanding our global testing capacity is critical to preventing and containing pandemics1-9. Accordingly, accessible and adaptable automated platforms that in decentralized settings perform nucleic acid amplification tests resource-efficiently are required10-14. Pooled testing can be extremely efficient if the pooling strategy is based on local viral prevalence15-20; however, it requires automation, small sample volume handling and feedback not available in current bulky, capital-intensive liquid handling technologies21-29. Here we use a swarm of millimetre-sized magnets as mobile robotic agents ('ferrobots') for precise and robust handling of magnetized sample droplets and high-fidelity delivery of flexible workflows based on nucleic acid amplification tests to overcome these limitations. Within a palm-sized printed circuit board-based programmable platform, we demonstrated the myriad of laboratory-equivalent operations involved in pooled testing. These operations were guided by an introduced square matrix pooled testing algorithm to identify the samples from infected patients, while maximizing the testing efficiency. We applied this automated technology for the loop-mediated isothermal amplification and detection of the SARS-CoV-2 virus in clinical samples, in which the test results completely matched those obtained off-chip. This technology is easily manufacturable and distributable, and its adoption for viral testing could lead to a 10-300-fold reduction in reagent costs (depending on the viral prevalence) and three orders of magnitude reduction in instrumentation cost. Therefore, it is a promising solution to expand our testing capacity for pandemic preparedness and to reimagine the automated clinical laboratory of the future.

Multiple-Valued Logic Modelling for Agents Controlled via Optical Networks

The methods of data verification are discussed, which are intended for the distant control of autonomous mobile robotic agents via networks, combining optical data links. The problem of trust servers is considered for position verification and position-based cryptography tasks. In order to obtain flexible quantum and classical verification procedures, one should use the collective interaction of agents and network nodes, including some elements of the blockchain. Multiple-valued logic functions defined within discrete k-valued Allen–Givone algebra are proposed for the logically linked list of entries and the distributed ledger, which can be used for distant data verification and breakdown restoration in mobile agents with the help of partner network nodes. A distributed ledger scheme involves the assigning by distant partners of random hash values, which further can be used as keys for access to a set of distributed data storages, containing verification and restoration data. Multiple-valued logic procedures are simple and clear enough for high-dimensional logic modelling and for the design of combined quantum and classical protocols.

Millimeter Wave Wireless Assisted Robot Navigation With Link State Classification

The millimeter wave (mmWave) bands have attracted considerable attention for high precision localization applications due to the ability to capture high angular and temporal resolution measurements. This paper explores mmWave-based positioning for a target localization problem where a fixed target broadcasts mmWave signals and a mobile robotic agent attempts to capture the signals to locate and navigate to the target. A three-stage procedure is proposed: First, the mobile agent uses tensor decomposition methods to detect the multipath channel components and estimate their parameters. Second, a machine-learning trained classifier is then used to predict the <i>link state</i>, meaning if the strongest path is line-of-sight (LOS) or non-LOS (NLOS). For the NLOS case, the link state predictor also determines if the strongest path arrived via one or more reflections. Third, based on the link state, the agent either follows the estimated angles or uses computer vision or other sensor to explore and map the environment. The method is demonstrated on a large dataset of indoor environments supplemented with ray tracing to simulate the wireless propagation. The path estimation and link state classification are also integrated into a state-of-the-art neural simultaneous localization and mapping (SLAM) module to augment camera and LIDAR-based navigation. It is shown that the link state classifier can successfully generalize to completely new environments outside the training set. In addition, the neural-SLAM module with the wireless path estimation and link state classifier provides rapid navigation to the target, close to a baseline that knows the target location.

Reinforcement learning-based dynamic obstacle avoidance and integration of path planning.

Deep reinforcement learning has the advantage of being able to encode fairly complex behaviors by collecting and learning empirical information. In the current study, we have proposed a framework for reinforcement learning in decentralized collision avoidance where each agent independently makes its decision without communication with others. In an environment exposed to various kinds of dynamic obstacles with irregular movements, mobile robot agents could learn how to avoid obstacles and reach a target point efficiently. Moreover, a path planner was integrated with the reinforcement learning-based obstacle avoidance to solve the problem of not finding a path in a specific situation, thereby imposing path efficiency. The robots were trained about the policy of obstacle avoidance in environments where dynamic characteristics were considered with soft actor critic algorithm. The trained policy was implemented in the robot operating system (ROS), tested in virtual and real environments for the differential drive wheel robot to prove the effectiveness of the proposed method. Videos are available at https://youtu.be/xxzoh1XbAl0.

Navigation of a wheeled mobile robotic agent using modified grey wolf optimization controller

PurposeSmooth and autonomous navigation of mobile robot in a cluttered environment is the main purpose of proposed technique. That includes localization and path planning of mobile robot. These are important aspects of the mobile robot during autonomous navigation in any workspace. Navigation of mobile robots includes reaching the target from the start point by avoiding obstacles in a static or dynamic environment. Several techniques have already been proposed by the researchers concerning navigational problems of the mobile robot still no one confirms the navigating path is optimal.Design/methodology/approachTherefore, the modified grey wolf optimization (GWO) controller is designed for autonomous navigation, which is one of the intelligent techniques for autonomous navigation of wheeled mobile robot (WMR). GWO is a nature-inspired algorithm, which mainly mimics the social hierarchy and hunting behavior of wolf in nature. It is modified to define the optimal positions and better control over the robot. The motion from the source to target in the highly cluttered environment by negotiating obstacles. The controller is authenticated by the approach of V-REP simulation software platform coupled with real-time experiment in the laboratory by using Khepera-III robot.FindingsDuring experiments, it is observed that the proposed technique is much efficient in motion control and path planning as the robot reaches its target position without any collision during its movement. Further the simulation through V-REP and real-time experimental results are recorded and compared against each corresponding results, and it can be seen that the results have good agreement as the deviation in the results is approximately 5% which is an acceptable range of deviation in motion planning. Both the results such as path length and time taken to reach the target is recorded and shown in respective tables.Originality/valueAfter literature survey, it may be said that most of the approach is implemented on either mathematical convergence or in mobile robot, but real-time experimental authentication is not obtained. With a lack of clear evidence regarding use of MGWO (modified grey wolf optimization) controller for navigation of mobile robots in both the environment, such as in simulation platform and real-time experimental platforms, this work would serve as a guiding link for use of similar approaches in other forms of robots.

Toward Robotic Cognition by Means of Decision Tree of Deep Neural Networks Applied in a Humanoid Robot

One of the challenges of Deep Learning research is to develop algorithms for mobile robotic agents that operate in uncontrolled environments, in which dynamic changes and limited processing power are common restrictions. A common solution is to develop separate vision and decision modules, so that the former is based on deep neural network architectures and the latter is based on rules, and then interconnect them. The drawback of this solution is that the modules need to exchange high-level information about the objects in the scene, which are usually the positions of all objects in the scene, and this is computationally expensive. To address this problem, this paper presents a Decision Tree of Deep Neural Networks (DT-DNN) that aims to perform end to end—from image to decision—processing, and, thus, eliminating the need for quantitative and relational information about the image. This model is composed of smaller and more specialized modular DNNs, thus solving the trade-off between performance and inference time. Experiments were carried out using a real robot in the RoboCup Humanoid League domain in a soccer field, and also in simulation. We compared DT-DNN with several traditional DNN architectures. From the results, it is possible to conclude that the use of the DT-DNN made the system simpler and more robust, with fewer parameters to be adjusted, reducing the time spent with inference and also increasing the performance when compared to the traditional approach.

A distributed deployment algorithm for communication coverage in wireless robotic networks

Wireless Robotic Networks (WRNs), composed of numerous mobile robotic agents with the ability of moving, computing, sensing, and communicating, are able to provide wireless communication services and thus implement complicated tasks for user equipments. In order to guarantee the performance of coverage rate and ensure providing the services timely and effectively, mobile robotics in WRNs are usually deployed flexibly and quickly. However, coverage overlaps and energy redundancy may be generated by excessive deployed agents. In order to provide maximum coverage area with a minimum number of agents, we study the 3-dimensional coverage deployment problem in WRNs and propose a distributed deployment algorithm. Firstly, we give the scenario model, communication model, and coverage model to define the 3-dimensional coverage problem. Secondly, we propose a distributed coverage deployment algorithm that can set redundant agents to idle mode iteratively. Herein, we decouple the coverage deployment problem in the altitude and horizontal dimensions without any loss of optimality. On the one hand, this algorithm can find the optimal altitude for agents mathematically. On the other hand, this algorithm contains a local deployment algorithm based on Particle Swarm Optimization (PSO) which is used for a particular active agent to find a better position with a larger local coverage area. In particular, the local coverage area is obtained depending on the Voronoi Diagram (VD). Our proposed algorithm is distributed which only requires local information. Finally, performance evaluation are given in three aspects, which demonstrate the effectiveness of the proposed distributed deployment algorithm.

PyDSLRep: A domain-specific language for robotic simulation in V-Rep.

Calculating forward and inverse kinematics for robotic agents is one of the most time-intensive tasks when controlling the robot movement in any environment. This calculation is then encoded to control the motors and validated in a simulator. The feedback produced by the simulation can be used to correct the code or to implement the code can be implemented directly in the robotic agent. However, the simulation process executes instructions that are not native to the robotic agents, extending development time or making it preferable to validate the code directly on the robot, which in some cases might result in severe damage to it. The use of Domain-Specific Languages help reduce development time in simulation tasks. These languages simplify code generation by describing tasks through an easy-to-understand language and free the user to use a framework or programming API directly for testing purposes. This article presents the language PyDSLRep, which is characterized by the connection and manipulation of movement in mobile robotic agents in the V-Rep simulation environment. This language is tested in three different environments by twenty people, against the framework given by V-Rep, demonstrating that PyDSLRep reduces the average development time by 45.22%, and the lines of code by 76.40% against the Python framework of V-Rep.

Fuzzy rule-based environment-aware autonomous mobile robots for actuated touring

The involvement of computer-programmed autonomous mobile robots in real-time activities is emerging in the recent years. The actuation and interaction of the robots are controlled through optimized high-level programming to respond to environmental factors. Such robots require an optimized touring plan with a better response to understand the inherent conditions. In this article, fuzzy rule-based optimization for actuated touring (FOAT) is presented. FOAT is responsible for balancing the actuation and response of the mobile robot agents in touring and path exploration. The touring and self-decision analysis of the programmed robots is improved through FOAT by adapting the environmental conditions and then constructing rules for response. Different from the functions of line-based or other robot touring, the proposed actuated touring frames decisive rules based on the varying inputs. With the framed rules, the touring process of the robot is modified to achieve the best solution instantly adaptive to the environment. The performance of FOAT is verified through experiments and is then analyzed using the metrics: touring cost, obstacles hit ratio, time-lapse, and tour length.

Jointly Improving Parsing and Perception for Natural Language Commands through Human-Robot Dialog

In this work, we present methods for using human-robot dialog to improve language understanding for a mobile robot agent. The agent parses natural language to underlying semantic meanings and uses robotic sensors to create multi-modal models of perceptual concepts like red and heavy. The agent can be used for showing navigation routes, delivering objects to people, and relocating objects from one location to another. We use dialog clari_cation questions both to understand commands and to generate additional parsing training data. The agent employs opportunistic active learning to select questions about how words relate to objects, improving its understanding of perceptual concepts. We evaluated this agent on Amazon Mechanical Turk. After training on data induced from conversations, the agent reduced the number of dialog questions it asked while receiving higher usability ratings. Additionally, we demonstrated the agent on a robotic platform, where it learned new perceptual concepts on the y while completing a real-world task.

Intelligent Space environment for ethorobotics

This paper presents a setup using a mobile robot agent in an intelligent space. The proposed concept based on a holonomic driving system mount on the robot. Due to the driving system different moving behaviours and path planning algorithms can be tested. To examine the robot movements and provide the needed support the robot agent was placed in a Motion capture system. The special environment around the robot is a motion track system which can track the movements of agents in its space via markers.

Symmetry and reduction in collectives: low-dimensional cyclic pursuit.

We investigate low-dimensional examples of cyclic pursuit in a collective, wherein each agent employs a constant bearing (CB) steering law relative to exactly one other agent. For the case of three agents in the plane, we characterize relative equilibria and pure shape equilibria of associated closed-loop dynamics. Re-scaling time yields a reduction of phase space to two dimensions and effective tools for stability analysis. Study of bifurcation of a family of collinear equilibria dependent on a single CB control parameter reveals the presence of a rich collection of trajectories that are periodic in shape and undergo precession in physical space. For collectives in three dimensions, with an appropriate notion of CB pursuit strategy and corresponding steering law, the two-agent case proves to be explicitly integrable. These results suggest control schemes for small teams of mobile robotic agents engaged in area coverage tasks such as search and rescue, and raise interesting possibilities for behaviour in biological contexts.

Design of Navigation Control Architecture for an Autonomous Mobile Robot Agent

Background/Objectives: To design a navigation control architecture which attempts to identify optimal path for a mobile robot agent to move from initial position to target position, in an environment where multiple path exists. Methods/ Statistical Analysis: The design of navigation control architecture determines how all the components are integrated and it is completely depends on its requirement and since, there is no standard reference architectures exists, we propose an architecture for mobile robot system, with three-tier user centric layer approach with hybrid methodology. The integration and interaction between the three layers are both centralized as well as distributed and, choice of selecting the path planner mechanism differs from existing system. Findings: In this paper, various existing robot architectures were reviewed in high level. A Hybrid methodology is used to adopt along with dynamic modelling approach for motion planning system and geometric modelling approach for path planning system, to suit the design complexity. The Path planner control system model proposed in this paper focuses much on agent based path planning system and motion planning system, which plays critical role in controlling the movement of a non-holonomic robot system. Application/Improvements: The efficiency of the proposed architecture is yet to be evaluated by applying this system in real time application

Performance comparison of novel WNN approach with RBFNN in navigation of autonomous mobile robotic agent

This paper addresses the performance comparison of Radial Basis Function Neural Network (RBFNN) with novel Wavelet Neural Network (WNN) of designing intelligent controllers for path planning of mobile robot in an unknown environment. In the proposed WNN, different types of activation functions such as Mexican Hat, Gaussian and Morlet wavelet functions are used in the hidden nodes. The neural networks are trained by an intelligent supervised learning technique so that the robot makes a collision-free path in the unknown environment during navigation from different starting points to targets/goals. The efficiency of two algorithms is compared using some MATLAB simulations and experimental setup with Arduino Mega 2560 microcontroller in terms of path length and time taken to reach the target as an indicator for the accuracy of the network models.

A Distributed Computing Algorithm for Deployment of Mobile Robotic Agents with Limited Sensing Ranges

In this paper, the authors present a distributed computing algorithm for an application of deploying a group of mobile robotic agents in a convex region. Each agent can be treated as an embedded system with its own sensors and actuators. Under the assumption that agents have limited sensing capabilities, the distributed computing algorithm is designed to make agents cover the convex region so that each member's individual region, the space around them, is optimized. Each agent must accommodate its kinematic constraints and can only exchange information locally based on the range of the sensors equipped. The proposed algorithm utilizes Voronoi partitions to create individual subregions and directs each member toward the centroid of their subregion. The Voronoi partitions are updated with each iteration of the algorithm. Simulation results are provided to validate the proposed algorithm.

Embedded GPU and multicore processors for emotional-based mobile robotic agents

Control architectures based on emotions are becoming promising solutions for the implementation of future robotic systems. The basic controllers of this architecture are the emotional processes that decide which behaviors the robot must activate to fulfill the objectives. The number of emotional processes increases (hundreds of millions/s) with the complexity level of the application, limiting the processing capacity of a main processor to solve the complex problems. Fortunately, the potential parallelism of emotional processes permits their execution in parallel, hence enabling the computing power to tackle the complex dynamic problems. In this paper, Graphic Processing Unit (GPU), multicore processors and single instruction multiple data (SIMD) instructions are used to provide parallelism for the emotional processes. Different GPUs, multicore processors and SIMD instruction sets are evaluated and compared to analyze their suitability to cope with robotic applications. The applications are set-up taking into account different environmental conditions, robot dynamics and emotional states. Experimental results show that, despite the fact that GPUs have a bottleneck in the data transmission between the host and the device, the evaluated GTX 670 GPU provides a performance of more than one order of magnitude greater than the initial implementation of the architecture on a single core. Thus, all complex proposed application problems can be solved using the GPU technology in contrast to the first prototype where only 55% of them could be solved. Using AVX SIMD instructions, the performance of the architecture is increased in 3.25 times in relation to the first implementation. Thus, from the 27 proposed applications about 88.8% are solved. In the case of the SSE SIMD instructions, the performance is almost doubled and the robot could solve about 74% of the proposed application problems. The use of AVX and SSE SIMD instructions provides almost the same performance as a quad- and a dual-core, respectively, with the advantage that they do not add any additional hardware cost.

Robot Guided Crowd Evacuation

The congregation of crowd undoubtedly constitutes an important risk factor, which may endanger the safety of the gathered people. The solution reported against this significant threat to citizens safety is to consider careful planning and measures. Thereupon, in this paper, we address the crowd evacuation problem by suggesting an innovative technological solution, namely, the use of mobile robot agents. The contribution of the proposed evacuation system is twofold: (i) it proposes an accurate Cellular Automaton simulation model capable of assessing the human behavior during emergency situations and (ii) it takes advantage of the simulation output to provide sufficient information to the mobile robotic guide, which in turn approaches and redirects a group of people towards a less congestive exit at a time. A custom-made mobile robotic platform was accordingly designed and developed. Last, the performance of the proposed robot guided evacuation model has been examined in real-world scenarios exhibiting significant performance improvement during the crucial first response time window.

Multicore and FPGA implementations of emotional-based agent architectures

Control architectures based on Emotions are becoming promising solutions for the implementation of future robotic agents. The basic controllers of the architecture are the emotional processes that decide which behaviors of the robot must activate to fulfill the objectives. The number of emotional processes increases (hundreds of millions/s) with the complexity level of the application, reducing the processing capacity of the main processor to solve complex problems (millions of decisions in a given instant). However, the potential parallelism of the emotional processes permits their execution in parallel on FPGAs or Multicores, thus enabling slack computing in the main processor to tackle more complex dynamic problems. In this paper, an emotional architecture for mobile robotic agents is presented. The workload of the emotional processes is evaluated. Then, the main processor is extended with FPGA co-processors through Ethernet link. The FPGAs will be in charge of the execution of the emotional processes in parallel. Different Stratix FPGAs are compared to analyze their suitability to cope with the proposed mobile robotic agent applications. The applications are set up taking into account different environmental conditions, robot dynamics and emotional states. Moreover, the applications are run also on Multicore processors to compare their performance in relation to the FPGAs. Experimental results show that Stratix IV FPGA increases the performance in about one order of magnitude over the main processor and solves all the considered problems. Quad-Core increases the performance in 3.64 times, allowing to tackle about 89 % of the considered problems. Quad-Core has a lower cost than a Stratix IV, so more adequate solution but not for the most complex application. Stratix III could be applied to solve problems with around the double of the requirements that the main processor could support. Finally, a Dual-Core provides slightly better performance than stratix III and it is relatively cheaper.