RFID rack assembly development for mobile platform.

Randmaa, Merili ; Otto, Tauno ; Kuusik, Alar 等

1. INTRODUCTION

RFID (Radio Frequency Identification) technology is becoming increasingly prevalent as the price of the technology decreases. There are various applications where RFID technology can establish new opportunities and qualities, such as RFID-assisted indoor localization and communication (Miller, 2006). Common way to test and develop new technologies is a model experiment on robots. Axiomatic Design (AD) provides a framework to describe design objects and a set of axioms to evaluate relations between intended functions and means by which they are achieved. During AD client needs, functionality of the product, parametric design requirements and manufacturability are analysed, whereas it is possible to use the achieved data for further analysis and save it into knowledge base (Lossack & Grabowski, 2000). Instead of conventional the proactive product development integrates consumer requirements model to potential avenues of consumer input and to a list of consumer product evaluative criteria (May-Plumlee & Little, 2006).

The development of a decision support system will become practical in the future using an accumulated database consisting of each systemized database for a given axiomatic design module. This will reduce the required period of development, minimize failures and mistakes during the design and development steps, and provide instant cause analysis (Lee et al., 2007). The impact of design changes on the subsequent implementation processes has never been considered in concurrent engineering design (Xue et al., 2006).

Market of personal service robots is increasing (Fig. 1), therefore exists need for service robot accessories for educational and research institutions.

[FIGURE 1 OMITTED]

According to the projections for the period 2009-2012, the stock of service robots for professional use is forecasted to increase to some 49,000 units (World Robotics, 2009).

The case study example is developing RFID antenna rack assembly for iRobot Roomba 530 on request of ELIKO.

ELIKO is currently focused on SSTS technology (Smart Space Technologies and Services), aiming to develop an open knowledge environment for self-configurable, low cost and robust robot swarms useable in everyday applications.

2. SERVICE ROBOT DEVELOPMENT

Robot Swarm (RS) solution is mainly targeting low service cost robotics for dynamic environments offering hardware independent software for robot knowledge exchange in case of weak dependability requirements. The projects done so far did not end up with any directly marketable component. The main results are in segment of free software. In current paper AD is analysed regarding e-business possibilities in the area of niche products like swarm robots, including also hardware solutions.

Developed robot swarm (Tammet et al., 2008) possesses a higher intelligence collectively than each member of the swarm independently. The robots divide a large scale into individual tasks for increasing functionality of the swarm. The system is self learning from the experience of individual swarm members via the global and local knowledge base. The technologies used are decentralized sensing, RFID (Radio Frequency Identification) and wireless communication. Present solution is using Ultra High Frequency (900 MHz) passive tags--tags use energy emitted by reader antenna(s). For robot application it is essential to achieve communication range of 0.5-1 m. Such distance is achievable in combination of 0.5 W reader (transmitter), which was technical limitation and selected 6 dBi "patch type" antenna. This particular antenna type is a carefully chosen trade off between mechanical and electrical properties.

3. MORPHOLOGICAL ANALYSIS

The methods to be used are morphological matrix, evaluation matrix, kinematic analysis and CosmosXpress stress analysis. Morphological matrix enables to create N different solutions N (Eq. 1), where mi is the number of partial solutions, for partial function i.

N = [m.sub.1] * [m.sub.2] ... [m.sub.n-1] * [m.sub.n] (1)

After combining different materials with different constructions, working principles and adjustment styles, we have many solutions to choose from. All the options need to be analyzed carefully, using kinematic analysis. Every requirement needs to be scaled by the importance beforehand to provide the most objective result from the evaluation matrix.

Present design is using two RFID antennas. Radio field reception angle of used custom antenna is around [+ or -]30[degrees]. Multiplexing between two antennas allows increasing 'visible' range of robot to [+ or -]50[degrees] from moving direction. Additionally, this configuration allows moving precisely towards a tag when it is detected with both antennas- overlapping sensitive area of both antennas is 10-15[degrees].

Since it is extremely difficult to realize 900 MHz patch type antennas with identical radio field sensitivity curves, the mounted antenna position has to be adjustable.

A RFID antenna rack assembly is needed for research on mobile iRobot Roomba 530 platform. iRobot Roomba is a widely used robot platform for research and development of various new technologies and usabilities. Developed rack assembly can also be used in alternative Roomba RFID projects.

The cusomers' needs to be translated into technical requirements, included the frame fitting on robot platform; the frame requirements to hold two RFID antennas and a PCB, whereas the angle of the antennas must be adjustable; the antennas must stand 30 mm above Roombas case; the frame should not conduct electricity; the frame needs to be easily detached from Roomba; the frame needs to be designed for small- and medium production.

4. OPTIMISING RFID RACK ASSEMBLY

Evaluation matrix displays that the best option is to cut the frame out of plastic sheets and glue them together. The most stable was a triangle-shape placement of the elements, which also reduces frames bending on sides when accelerating. Antenna adjustment is done by loosening the bolts, choosing another angle for the antennas and tightening the bolts again. Rack assembly is attached to Roomba by using Velcro.

Preliminary construction is designed, using 3D drawing program Solidworks and stress analysis made in CosmosXpress. The results showed that the frame can be optimized (Fig. 2). Workbench Datron M7HP was used to mill the details, so there was only few restrictions on the shape of the details. The details are glued together, butterfly nuts are used to adjust the angle of the antennas easily. Considering the price level of Roomba accessories at Roomba Shop, the recommended sale-prize is generating profit if manufacturing orders are repeating. So far the dominating sales tend to be freeware/GPL. The RS results with knowledge collection/exchange solution for robotics communities that does not have clear competitor. There is only small amount of RS like software solutions available for (non-microcontroller based) wide range robotics applications, e.g Microsoft Robotics Studio, Player, Ocra (Brooks, 2007). However, for actual takeoff of services to collect information payed services seems inappropriate, at least in the beginning, since the benefit to individual repository user is small. The sophisticated swarm robotics development kit (Fig. 3) with unit price around a thousand EUR should have market for at least some thousand devices in Europe and U.S., and is manufactured on demand basis

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

5. CONCLUSIONS

RFID antenna rack assembly was designed to support scientific project Roboswarm and other similar projects based on service robot platform. Due to its reasonable cost, open source programs and universal sensors, iRobot Roomba 530 mobile robot platform can be successfully used for the basis. The frame holds two ultra high frequency RFID antennas and a PC104 cirquit board. As mentioned, sale of such kits is supporting action for promoting robot swarm software technology. Further research needs to be done on how exactly to market these assemblies. The prototype solution is realized in Competence Centre ELIKO.

6. ACKNOWLEDGEMENTS

This work was supported by the Estonian Scientific Foundation grant ETF7852 and Estonian Ministry of Education and Research Project SF0140113Bs08.

7. REFERENCES

Brooks, A.; Kaupp, T.; Makarenko, A.; Williams, S. & Oreback, A. (2007). Orca: a component model and repository, Software Engineering for Experimental Robotics. Springer Tracts in Advanced Robotics, 30, pp 231-251

Lee, K.S.; Cha,S.W. & Koo, S.M. (2007). Applying Axiomatic Design to Improve Thermoforming Multilayered Plastic Container, Polymer-Plastics Technology and Engineering, 46, pp 781-788

Lossack, R. & Grabowski, H. (2000). The axiomatic approach in the universal design theory, Proceedings of ICAD2000 First International Conference on Axiomatic Design, Cambdridge, MA--June 21-23, 2000, ICAD005

May-Plumlee, T. & Little, T.J. (2006). Proactive product development integrating consumer requirements, International Journal of Clothing Science and Technology, Vol. 18 No. 1, 2006, pp 53-66

Miller, L.E.; Wilson, O.F.; Bryner, N.P.; Francis, M.H.; Guerrieri, J.R.; Stroup, D.W. & Klein-Berndt, L. (2006). RFID-Assisted Indoor Localization and Communication for First Responders, Proc. of the Int. Symposium on Advanced Radio Technologies, pp 1-6

Tammet, T.; Vain, J.; Kuusik, A. & Puusepp, A. (2008). RFID based Communication for a Self-Organizing Robot Swarm, Proceedings: Second IEEE International Conference on Self-Adaptive and Self-Organizing Systems SASO2008, Venice, Italy October 20-24, 2008. IEEE Computer Society, 2008, pp 45-54

Xue, D.; Cheing, S.Y. & Gu, P. (2006). Configuration design considering the impact of design changes on downstream processes based upon the axiomatic design approach, Journal of EngineeringDesign,17, 6, pp 487--508

World Robotics. (2009). Professional service robots are establishing themselves, Available from: http://www.worldrobotics.org/, Accessed on: 2009-09-30