Multidisciplinary Design of a Micro-USV for Reentry Operations

Abstract

This paper presents the robust multidisciplinary design of a small scale unmanned space vehicle for re-entry operations. Uncertainties on the aerodynamic forces, thermal flux and characteristics of the thermal protection material are incorporated into the vehicle model. A Monte-Carlo sampling procedure is used to compute the mean and variance of the maximum heat flux and internal temperature. Then a population-based multi-objective optimization algorithm of the EDA type (Estimation of Distribution Algorithms) searches for geometries that minimize the mean value of the maximum heat flux, mean value of the maximum internal temperature, and the weighted sum of their variance. The EDA is hybridized with a direct transcription method for optimal control problems. The evolutionary part handles the shape parameters of the vehicle and the uncertain objective functions, while the direct transcription method generates an optimal control profile for the re-entry trajectory. A multi-fidelity approach is adopted during the optimization process to estimate the correct aerodynamic properties of the vehicle. Initially a low-fidelity analytical model, fitted on a wave-rider type of vehicle, is used to train an artificial neural network (ANN) which approximates the aerodynamic forces. Through the evolution a mix of analytical and CFD high-fidelity calculations are used to update the ANN. CFD runs progressively become the main source of updates for the ANN till, close to convergence of the EA, the analytical model is almost completely dropped. Some preliminary results show the achievable performance of such a small spacecraft and the requirements in terms of thermal protection materials.

Multiobjective Evolutionary Algorithms in Aeronautical and Aerospace Engineering

“…The authors compared the use of the NSGA-II to a reference (nearly optimal) solution, and concluded that their hybrid approach was very useful for estimating the optimum and for narrowing down the search in the presence of uncertainties. 8) Minisci et al [111] dealt with the robust multidisciplinary preliminary design of a small-scale unmanned space vehicle (USV), which was planed to be used for space re-entry operations. In this case, the simultaneous optimization of both, the spacecraft shape, and its trajectory control profile, are required.…”

“…This is to say that a MOEA is used to find a set of good solutions, perhaps at a coarse granularity (e.g., without considering all the decision variables), which are further improved using a local search engine (gradient-based techniques are normally used for this sake). For example, in [111], the MOEA is used at an upper level, with a subset of the decision variables and without incorporating any constraints, while the constraints and all the decision variables are considered and solved at a lower level, in which a gradientbased optimization process is used to find feasible solutions. Although memetic MOEAs have existed for several years in the specialized literature [50], the development of specific MOEA-based approaches that properly combine a global and a local search scheme in an efficient and effective way when dealing with space applications, is still an open research area.…”

Robust design of a re-entry unmanned space vehicle by multi-fidelity evolution control

“…In [19] the authors describe a similar approach applied to the design of a small scale USV, taking into account the availability of last generation thermal protection systems (TPS) based on ultra-high temperature ceramic materials (UHTC) [30,27]. Here the integrated technique is applied to the design of a medium scale USV, considering a more complex implementation, with two different trajectory optimisations: different control laws allow the same vehicle to follow a) the path that minimises the maximum heat flux, and b) the one maximising the orthodromic distance, during the re-entry phase.…”

“…Here the integrated technique is applied to the design of a medium scale USV, considering a more complex implementation, with two different trajectory optimisations: different control laws allow the same vehicle to follow a) the path that minimises the maximum heat flux, and b) the one maximising the orthodromic distance, during the re-entry phase. As in [19], here the shape of the vehicle is derived from on an ideal waverider configuration ( [29,15,28]), and its geometry is modified in order to introduce more realistic rounded edges.…”

Multidisciplinary Design Optimization of Aerospace Transportation Systems

“…In [9] the authors describe a similar approach applied to the design of a small-scale USV, taking into account the availability of last-generation thermal protection systems (TPS) based on ultra-high temperature ceramic materials (UHTC) [10,11]. Here the integrated design technique is applied to the design of a larger scale USV, and a different treatment of the optimal control component is introduced: a dual trajectory optimization loop generates control law solutions that minimize the maximum heat flux or maximize the orthodromic distance during the reentry phase.…”