Robust design of a robot wooden link
This thesis is in the framework of a French Région Pays de la Loire project named RobEcolo. The main goal of RobEcolo is to lower the environmental impact of industrial robots without decreasing their performance in terms of accuracy. One of the investigated fields of research is to replace metallic materials in robots by biosourced renewable materials, such as the wood.
Wood was used in the past for designing industrial machines (see for example the Jacquard loom designed in the 19th century) but was progressively replaced by metallic (and later the composite) materials. However it is still used for the design of buildings due to its high payload-to-weight ratio and its low cost. Nowadays, designing machines with wood is limited to mockup, prototyping  or entertainment. It is also used in machines in which the use of wooden element does not impact the machine accuracy, e.g. chassis of cars. Indeed, Industry does not trust such a material whose performance varies with temperature, humidity, etc.
A recent attempt to introduce wood in robot manufacture has been published in(Fig. B1-1). The results showed that the approach was valid enough to compete with usual non-renewable materials. However, this study missed crucial issues, among which: the wood performance / dimensions will vary with the surrounding atmospheric conditions ; thus, new robot design issues appear: How to be sure that the design process is robust wrt wood variability, i.e. that a wooden robot can be accurate, stiff, etc., even if wood properties vary ?
The aim of the present master thesis is to start investigation in this field. The idea is to show that wood variability will be overcome / decreased by searching for a proper combination of (i) wood type  (tree species, raw wood / composite, etc.), (ii) chemical treatment  (compromise pollution / performance stability) and (iii) link shaping (max. stiffness, low mass, etc.). Obviously, this cannot be enough and additional approached should be envisaged, but the thesis will be limited to the mentioned aspects.
Proposed work plan:
This thesis is in the continuation of the work of Mr. A. Nayak done during his master thesis in 2015 who optimized a wooden robot element in terms of inertia for one given loading. We need to go further, i.e. not to consider only the static deformation of the link, but also the natural frequencies. The main issue is to obtain a robust design for the link, i.e. a design whose performance in terms of inertia and vibrations which is as less sensitive as possible to the variation of the wood mechanical properties.
To obtain that, the student must first, after having read and understand the work done by Mr. Nayak, start modelling the natural frequencies of a robot link, and then find (with the help of his/her supervisors) the best optimization algorithm which can ensure the convergence towards the global minima representing the Pareto front.