Inside Unmanned Systems

AUG-SEP 2016

Inside Unmanned Systems provides actionable business intelligence to decision-makers and influencers operating within the global UAS community. Features include analysis of key technologies, policy/regulatory developments and new product design.

Issue link: http://insideunmanned.epubxp.com/i/720234

Contents of this Issue

Navigation

Page 53 of 59

54 unmanned systems inside August/September 2016 ugust/Septe August/September 2016 August/September 2016 be August/September 2016 0 August/September 2016 6 August/September 2016 AIR UAS RELIABILITY F igure 3 is a block diagram of how the servo fault can be detected before re- configuration. This figure shows an algorithm for model based fault diagnosis. Model based fault diagnosis is defined as "the determination of faults of a system from the comparison of available system measurements with a priori information represented by the system's mathematical model, through genera- tion of residual quantities and their analysis." 3 Stated differently, the output of any servo in response to a pilot's command can be modeled in the laboratory during the design phase of the UAS control system. Note the term pilot input in the case of Figure 3 is taken to mean either a human operator's or the control sys- tem's input. Now if the response of the servo is different from what we expect, we know a fault has occurred and it is time to reconfigure the control system to deal with the faulted servo. Hence the essence of this type of fault de- tection process lies in accurately modeling our system, here a servo actuator. This accurate servo model is the key ingredient in developing the residual filter shown in Figure 3. A servo model can be obtained by performing system identification on input-output data. 4 For the work described here a dedicated bench top ex- perimental test bed for actuator analysis was developed. This is a useful setup that allows us to analyze the response of both healthy and faulty actuators off line. Setting up such a char- acterization facility is not trivial as it requires replicating aerodynamic loading conditions on the benchtop. by Inchara Lakshminarayan, Raghu Venkataraman, Daniel Ossmann, Peter Seiler and Demoz Gebre-Egziabher DESIGNING RELIABILITY INTO SMALL UAS AVIONICS: part 2 Introduction If current projections of the utility of small Unmanned Aircraft Systems (UAS) are correct, then they will be common place sights in urban and rural skies. They will be in the suburban skies delivering packages to homes or as couriers in the urban downtown areas. In rural settings they will be around as precision agriculture platforms or for remote inspection of infrastructure such as power lines or railroad tracks. The purpose of this article is to explore, albeit brief ly, one implication UAS ubiquity has this article is to explore, albeit brief ly, one implication UAS ubiquity has on the design of guidance, navigation and control (GNC) systems of small aerial vehicles (class 1 and 2 UAS). For convenience and without a loss of generality, let us anchor our dis- cussion to one particular application of UAS: package delivery in urban settings. A small UAS used in this application will be expected to navi- gate in and above urban canyons. It will have to do this without colliding with buildings, power lines or other UAS. Faults in the guidance, naviga- tion or control systems can lead to collision. These faults can either be the result of hardware failure or the result of software and algorithmic shortcomings. Any engineered system will not be fault-free and, thus, it is unrealistic to expect 100% mishap free operations of UAS. However, we can and should try to minimize the likelihood of GNC system faults that can lead to collisions to a level considered acceptable by society at large. To do this we need to answer, at least, the following three ques- tions: (1) What is the maximum acceptable failure rate?, (2) How do we design GNC systems to meet such a failure rate requirement?, and (3) How do we prove a given GNC system design meets the failure rate re- quirement? These are not new questions and designers of GNC system used on manned aircraft or large aerospace systems (e.g., missiles, satel- lites, etc.) have addressed or dealt with them in the past. Therefore, before proceeding to answer these questions for UAS it might be instructive to see how these questions have been answered or dealt with by those who design GNC systems for manned aircraft. Detecting a Failed Servo KEY INSIGHTS The article describes the use of analytical redundancy to improve the reliability of low cost unmanned aircraft system (UAS). Fault monitoring systems are presented and tested for the actuation mechanism and navi- gation system. The reconfi guration of a UAS in the presence of system failures is successfully demonstrated using real fl ight tests.

Articles in this issue

Archives of this issue

view archives of Inside Unmanned Systems - AUG-SEP 2016