Emilie Poirson

User-centered design by genetic algorithms: Application to brass musical instrument optimization

This work presents an implementation of genetic algorithms (GAs) for a user-centered design of products. It describes at first a methodology for a user-centered design, based on the coupling between a subjective study to define desirable features and on objective study to find out the influencing objective variables. It relies on two domains that remain generally distinct: the design with a scientific approach (generally math-based) and the design with a sensory and perceptual approach (subjective). The methodology is presented on a particular product for which the perceived aspects are essential: a musical instrument (trumpet). Two types of study were carried out on a set of trumpets: firstly, a sensory study, which aim is to characterize the perception of the intonation of the instruments by musicians; secondly, an objective study, which consists in an objective description of the instruments by physical measurements (impedance). We correlated the intonation assessments data and the physical measurements, in order to deduct useful objective functions for the design of a new instrument, and to formulate the user-centered design problem as a multi-objective optimization problem. The paper presents next how GAs were implemented to solve the multi-objective optimization problem.

Integrated and interactive method for solving layout optimization problems

Having a significant impact on the design of many products and industrial systems, such as the subdivision of a ship, the layout of facilities in a plant or further still the assembly of parts of a mechanism, layout design optimization is at the heart of scientific issues. The design of an optimal layout solution is a critical and complex task due to the increasing demands of designers working on varied projects. This paper proposes an integrated approach to solve layout optimization problems, from the needs expressed by the designer to the creation of an ideal solution. This generic and interactive method is based on a design process divided into four steps: the description, the formulation, the solving of the problem, and the final decision. This process is based on a multiobjective modular optimization strategy that combines a genetic algorithm with local optimization modules. The method described in this paper is interactive because the designer participates in all processs steps. For example, in the final decision step, the approach includes an interactive environment in order to let the designer choose and improve an optimal solution according to his personal judgment and expertise. The global method is applied to an industrial problem which deals with the search for an optimal layout of facilities in a shelter.

Integrating User to Minimize Assembly Path Planning Time in PLM

Assembly and disassembly verification plays an important role in Product Lifecycle Management (PLM). Mass customization and personalized production require greater agility of industrial manufacturers, and consequently the time consumed in assembly processes (which often dominates costs) is of growing importance. Narrow passages which commonly exist in assembly tasks become a bottleneck in the assembly planning process reducing the time saved by using virtual assembly. In this paper, we present a novel interactive motion planning methodology for virtual assembly and disassembly operations. In our method, user’s manipulation with force feedback device in difficult scenario is liberated by relaxing collision constraints. The rough path is retracted by our random retraction method and then connected by BiRRT algorithm. For each user’s operation, a path successfully passing through the narrow passage is presented, or geometrical interference information in failed case is provided for motion re-planning and design modification. Significant improvement for trajectory planning of non-convex object is observed in challenging 2D scenarios with narrow passages.

Detecting Design Trends Using Perceptive Tests Based on an Interactive Genetic Algorithm

To avoid flops, the control of the risks in product innovation and the reduction of the innovation cycles require valid and fast customer’s assessments. An Interactive Genetic algorithm is proposed for eliciting user’s perceptions concerning the shape of a product, in order to stimulate creativity and detecting design trends. Interactive users’ assessment tests are conducted on virtual products, for capturing and analyzing users’ responses. The IGA is interfaced with a CAD software (CATIA V5) and allows the creation in real time of a set of parameterized designs, which are presented iteratively by a graphical interface to the user for evaluation. After a description of the IGA, we present a study on the convergence of the IGA, according to the tuning parameters of the algorithm and the size of the design problem. An experiment was carried out with a set of 20 users on a particular product, table glasses. We describe the implementation of the perceptive tests and an analysis of the results. These results show how the IGA can be used to elicit user perception and detecting design trends.Copyright

An Interactive-based approach to the Layout design optimization

Layout design plays an important role in the design and usability of many engineering products and systems. Because of the great complexity of most industrial layout problems, the decision of the acceptable layout is a hard and critical task since the special layout can have a significant consequence on the global performances. Thus, in order to propose to the designer an optimal spatial arrangement in a reasonable time, this chapter develops an interactive optimization strategy that is tested on the facilities layout problem of a shelter. This problem is innovative because it introduces the concept of space of accessibility

An interactive motion planning framework that can learn from experience

The accessibility verification of the assembly/disassembly plays an important role in the process of product design. In the last decade, the sampling based motion planners have been successfully applied to solve the accessibility verification. However, the narrow passage which is a common problem in the assembly tasks is still a bottleneck. Meanwhile, the requirement of perception and emotion assessment drives the interaction between users and automatic path planners in the virtual assembly process. In this paper, a curve matching method is used to explore the implicit relationship between the topological information of scenarios and the motion of objects, based on which an interactive motion planning framework that can learn from experience is constructed.Our framework consists of two main processes: a learning process and a motion generation process. In the former process, the motion segment (a part of motion path) and its related scenario segment (a part of workspace passed through by the object) are gathered, after an interactive motion planning process finds a collision-free motion path or reaches the conclusion of inaccessibility. According to the similarity between the skeletons of scenario segments, the gathered scenario segments and motion segments are organized by a hierarchical structure in the motion library. The latter process permits users to control only one point in the workspace for the selection of a new scenario, and then the similar scenarios are retrieved from the motion library, to help quickly detect the connectivity of the new scenario and generate a repaired motion path to guide users with feasible manipulations. We highlight the performance of our framework on a challenging problem in 2D, in which a non-convex object passes through a cluttered environment filled with randomly shaped and located non-convex obstacles. Maximal sphere sequence represents the scenario. The centerline of maximal sphere sequence represents the scenario skeleton.The motion of object has close relationship with the change of the skeleton and volume size of scenario.Motion learning is based on computing the similarity of scenarios by dynamic time warping (DTW).Scenario retrieval is highly efficient due to the hierarchical clustering in motion library.Releasing humans from rotation manipulation in complex assembly verification with learned motion experience.

Interactive Modular Optimization Strategy for Layout Problems

Layout design optimization has a significant impact in the design and use of many engineering products and systems. Real-world layout problems are usually considered as complex problems because of the geometry of components, the problem density and the great number of designers requirements. Solving these optimization problems is a hard and time consuming task. This paper proposes an interactive modular optimization strategy which allows the designer to find optimal solutions in a short period of calculation time. This generic strategy is based on a genetic algorithm, combined with specific optimization modules. These modules improve the global performances of the algorithm. This approach is adapted to multi-objective optimization problems and interactivity between the designer and the optimization process is used to make a final choice among design alternatives. This optimization strategy is tested on a real-world application which deals with the search of an optimal spatial arrangement of a shelter.

INTERACTIVE AND ON-LINE LEARNING SYSTEM FOR ASSEMBLY TASK MOTION PLANNING

This paper presents a novel interactive motion planning system for assembly/disassembly operations. Our system consists of three layers: interaction layer, learning layer and motion library layer. In interaction layer, user’s manipulation in difficult scenario is liberated by relaxing collision constraints. The resulting path is retracted and connected by random retraction method and BiRRT algorithm. A motion path which successfully passed through the narrow passage or information of geometrical interference in failed case is returned to user. In learning layer, motion primitives corresponding to prior similar scenario are selected by scenario comparison which is based on medial axis, and then transformed to generate new motions. Significant improvement for motion planning of non-convex object in challenging scenarios with narrow passages is obtained by interactive process. The introduction of learning mechanism can reduce global planning time and obtain experiential knowledge.© 2013 ASME

Learning from Product Users, a Sentiment Rating Algorithm

Social media gives new opportunities in customer survey and market survey for design inspiration with comments posted online by users spontaneously, in an oral-near language, and almost free of biases. This new source however has huge size and complexity of data needed to be processed. In this paper, we propose an automated way for processing these comments, using sentiment rating algorithm. Traps like negations, irony, smileys are considered in our algorithm. We validate it on the example of a commercial home theatre system, comparing our automated sentiment predictions with the one of a group of 15 test subjects, resulting in a satisfactory correlation.

Perceptive Exploration of Layout Designs Using an Interactive Genetic Algorithm

Layout design optimization has a significant impact in the design and use of many engineering products and systems, such as the subdivision of a ship, the layout of facilities in a plant or further still the assembly of parts of a mechanism. The search for an optimal layout configuration is a critical and complex task due to the increasing demands of designers working on varied projects. A layout optimization process is generally divided into different steps: the description, formulation and solving of the problem and the final decision. This process consists in writing an optimization problem that transform designer’s requirements into variables, constraints and objectives. Then, an optimization algorithm has to be used in order to search for optimal solutions that fit with product’s specifications. This paper focuses on the last step which consists, for the designer, in making a choice on the solutions generated by the optimization algorithm. This choice is made according to the global performances of the designs and also the personal judgment of the designer. This judgment is based on the expertise of the designer and the subjective requirements that could not be integrated on the formulation of the problem. This paper proposes a perceptive exploration method, based on an Interactive Genetic Algorithm (IGA), used to explore designs, taking into account the subjective evaluation of the designer. The objective of this method is to select an ideal solution that realizes the best trade-off between the quantitative and qualitative performance criteria. This interactive process is tested on an industrial layout application which deals with the search for an optimal layout of facilities in a shelter.Copyright