Steffen Junginger

A common wireless remote control system for mobile robots in laboratory

A common wireless remote control system based on standard APIs of robots is presented to enable a stable multi-robot transportation in distributed life science laboratories. This system consists of multi-robot board control centers (PCs), a remote server control center (PC), a wireless communication network and an infrared radio navigation module with ceiling passive landmarks. To let this system expand conveniently, the two-level Client/Server architecture is adopted, and a standard IEEE 802.11g wireless communication with TCP/IP protocol is utilized. An inside architecture is employed for signal sampling and controlling between robot board PCs and the robots hardware modules. An additional outside architecture is designed for higher remote commands between robot board PCs and remote server control PC. Two experiments in this study show that the simple ceiling landmark method is suitable for the robot indoor navigation with low costs, and this kind of remote control system can work effectively in large and distributed laboratory.

A Floyd-Dijkstra hybrid application for mobile robot path planning in life science automation

A new application for path planning is presented for mobile robot transportation in life science laboratories. In this application: (a) to decide the shortest paths for mobile tasks, a hybrid path planning strategy using two algorithms from the map theory is proposed. The Floyd algorithm is adopted to do an off-line path planning, and the Dijkstra algorithm is executed to decide an on-line alternative path when the Floyd based route is not available. Different to intelligence based planning methods (such as Artificial Neural Networks), the map theory based methods can definitely guarantee a global shortest path based on the prepared waypoints. This is important for a big life science environment; (b) besides the path planning issue, other main contents for mobile transportation are elaborated, including task dispatching, robot localization, communication architecture, XML-based command protocol, etc. Two experiments show that the proposed application and its developed software are effective for mobile robot transportation in life science laboratories.

A fast method for mobile robot transportation in life science automation

In modern life science laboratories, more and more complicated transportation tasks are needed among distributed automatic workbenches or laboratories. There are some special requirements for the kinds of transportation: high-accuracy; robust performance; economic cost and fast integration. In this paper a fast method is proposed to manage mobile robots for effective transportation in life science environments. The architecture of this method includes three components: the PMS for transportation request the Robot Remote Center (RRC) for transportation managing, and the Robot Board Center (RBC) for transportation executing. To include any kind and size of life science laboratories, this method adopts a standard TCP/IP network for data transmission, uses a series of extensible ceiling landmarks for robot indoor localization, utilizes a flexible map-based intelligent hybrid calculation for robot path planning, and enables collision avoidance by using a group of ultrasonic sensors and artificial potential field algorithm. An experiment in a real life science laboratory (celisca, Germany) shows that the proposed method meets all requirements of life science automation and has satisfactory performance.

An application of charging management for mobile robot transportation in laboratory environments

In life science laboratories, mobile robots are adopted to do transportation tasks among distributed automated islands or rooms. To use those mobile robots effectively, a number of technology issues have to be considered, such as path planning of transportation and charging, localization, communication, etc. In this paper an application of charging management for mobile robot transportation is presented. In this application: (a) to localize the mobile robots accurately and fast in life science environments, a method using ceiling landmarks is adopted; (b) to measure the power status from all running robots, a power module for the server and client sides is designed; (c) to let mobile robots go charging automatically, an automated charging station is utilized; and (d) to select the best installing positions for the charging stations, an intelligent method based on the Artificial Immune Algorithm (AIA) is proposed, which considers both of the distances among the expected working positions and their distribution of transportation tasks. A real case shows that the presented application is effective for mobile robot charging management in life science environments.

A floyd-genetic algorithm based path planning system for mobile robots in laboratory automation

In laboratory automation, more and more mobile robots have been employed for kinds of transportation among distributed automated ‘islands’. To direct those mobile robots effectively, in this paper a standard robot path planning system (RPPS) has been developed. In the RPPS: (a) to let the RPPS suit for any kind of mobile robots and any size laboratory, a TCP/IP based Client/Server architecture is adopted. A new robot distributed for a new laboratory can be added to the RPPS conveniently with a new IP; (b) an independent Robot Board Center (RBC) is developed for every robot. So even a robot loses its connection to a Robot Remote Center (RRC), it still can finish a given P2P task rightly; (c) to provide flexible robot paths for mobile tasks with different destinations, a map based path planning method is proposed. The map is comprised of waypoints which can be initialized quickly using a RBC. To calculate the shortest path for every P2P task, a hybrid calculation method based on Floyd algorithm and Genetic algorithm is designed; (d) to get low cost and extendable robot indoor localization for a large environment, a ceiling landmark localization is utilized; and (e) to avoid solving the complicated kinematic models for robot arm control, a way using a training arm is presented. Two experiments show that the proposed RPPS is effective for mobile robots in laboratory automation.

A Fast Approach to Arm Blind Grasping and Placing for Mobile Robot Transportation in Laboratories

This paper presents a fast approach to organizing arm grasping and placing manipulations for mobile robot transportation systems in life science laboratories. The approach builds a blind framework to realize the robot arm operations without integrating any other sensors or recognizing computation, but only adopting the robots existing on-board ultrasonic sensors originally installed for collision avoidance. To achieve high-precision indoor positioning performance for the proposed blind arm strategy, a hybrid method is proposed, including a StarGazer system for all laboratory environments and an ultrasonic sensor-based component for the local areas where the arm operations are expected. At the same time, two error-correcting algorithms are presented for the improvement of the high-precision localization and the selection of the robot arm operations. In addition, the architecture of all the robotic controlling centres and their key APIs are also explained. Finally, an experiment proves that the proposed blind strategy is effective and economically viable for the laboratory automation.

Biomek Cell Workstation A Variable System for Automated Cell Cultivation

Automated cell cultivation is an important tool for simplifying routine laboratory work. Automated methods are independent of skill levels and daily constitution of laboratory staff in combination with a constant quality and performance of the methods. The Biomek Cell Workstation was configured as a flexible and compatible system. The modified Biomek Cell Workstation enables the cultivation of adherent and suspension cells. Until now, no commercially available systems enabled the automated handling of both types of cells in one system. In particular, the automated cultivation of suspension cells in this form has not been published. The cell counts and viabilities were nonsignificantly decreased for cells cultivated in AutoFlasks in automated handling. The proliferation of manual and automated bioscreening by the WST-1 assay showed a nonsignificant lower proliferation of automatically disseminated cells associated with a mostly lower standard error. The disseminated suspension cell lines showed different pronounced proliferations in descending order, starting with Jurkat cells followed by SEM, Molt4, and RS4 cells having the lowest proliferation. In this respect, we successfully disseminated and screened suspension cells in an automated way. The automated cultivation and dissemination of a variety of suspension cells can replace the manual method.

Camera grids for laboratory automation

This paper presents a straightforward method for mapping a multi-camera system to inspect multiple analytical samples carried in a microtiterplate, a labware standard used in life-science laboratories today. The high number of small samples that are provided by microplates burden difficulties when designing an inspection system with a single camera. The major concern is the perspective that is unique for every sample, and only a limited number of samples can be imaged in a favourable perspective, that is directly from below or above the sample. A way to overcome this problem is to use multiple cameras. This work presents the integration of an arbitrary number of cameras to support microplates of all types and densities. It provides a mapping framework to deliver the subimages to the inspection algorithms and includes a one-step registration procedure that uses a registration plate with microplate dimensions. Furthermore, an estimation of the systems mapping uncertainty is given.

A Highly Flexible, Automated System Providing Reliable Sample Preparation in Element- and Structure-Specific Measurements

Life science areas require specific sample pretreatment to increase the concentration of the analytes and/or to convert the analytes into an appropriate form for the detection and separation systems. Various workstations are commercially available, allowing for automated biological sample pretreatment. Nevertheless, due to the required temperature, pressure, and volume conditions in typical element and structure-specific measurements, automated platforms are not suitable for analytical processes. Thus, the purpose of the presented investigation was the design, realization, and evaluation of an automated system ensuring high-precision sample preparation for a variety of analytical measurements. The developed system has to enable system adaption and high performance flexibility. Furthermore, the system has to be capable of dealing with the wide range of required vessels simultaneously, allowing for less cost and time-consuming process steps. However, the system’s functionality has been confirmed in various validation sequences. Using element-specific measurements, the automated system was up to 25% more precise compared to the manual procedure and as precise as the manual procedure using structure-specific measurements.

Human face orientation recognition for intelligent mobile robot collision avoidance in laboratory environments using feature detection and LVQ neural networks

In this paper, an approach on the intelligent mobile robot collision avoidance is proposed for the complex laboratory robot transportation process using the human face orientation recognition strategy. The proposed approach includes the contents as: (a) Measuring the face images of laboratory personnel by the adopted Microsoft Kinect sensors; (b) Processing the measured face images to recognize the face orientations which will be used to control the mobile robots in the collision avoidance; and (c) Building the Learning Vector Quantization (LVQ) Neural Networks to calculate and decide the face orientations based on the extracted face feature data. To select the best training algorithm for the LVQ model, a trail experiment is provided in the study. The results of the study show that: based on a standard laptop, the successful rate and the elapsed time of the proposed human face recognizing method are 99% and 3.17s, respectively. It means the proposed method can be applied in the mobile robot collision avoidance applications.