Spelling suggestions: "subject:"autorouting"" "subject:"autoroutier""
1 |
Automation of the design process of printed circuit boards : Determining minimum distance required by auto-routing softwareStröm, Simon, Qhorbani, Ali January 2018 (has links)
This thesis project aims to create an overview of new technologies in printed circuit board manufacturing which when automated could become part of an Industry 4.0 production flow. Potential design limits imposed by new technologies are then applied in the creation process of a minimum distance estimation function. The intended purpose of this function is to correctly estimate the minimum distance required for the auto-routing software FreeRouting to be able to successfully route between two components. This is achieved by using a brute-force attack to progressively decrease the distance between components using a bisectional approach to find the minimum distance at which the auto-routing software can still successfully route for a specific design. Using the results from this brute-force attack a couple of linear functions based on different base designs are created and then used to implement a minimum distance function. The minimum distance estimation function is then intended to be used to implement limits to how close components can be placed to each other in a printed circuit board design tool which purpose is to enable people with lesser knowledge of printed circuit boards to still be able to realize their design ideas. / Detta examensarbete ämnar skapa en överblick av nya tekniker inom mönsterkorts-tillverkning som när de automatiseras skulle kunna bli en del av ett Industri 4.0 produktionsflöde. Eventuella designbegränsningar som uppstår till följd av dessa tekniker kommer sedan appliceras i skapningsprocessen av en minsta avståndsfunktion. Syftet med denna funktion är att korrekt uppskatta det minimala avståndet som krävs för att auto-routing mjukvaran FreeRouting ska kunna dra ledningar mellan två komponenter. Detta görs genom en brute-force attackvinkel där avståndet mellan komponenter fortsätter minskas med bisektionsmetoden tills ett minsta avstånd hittats där auto-routing mjukvaran fortfarande kan dra ledningar för en specifik design. Genom användande av resultaten från denna brute-force attack skapas sedan ett par linjära funktioner baserade på olika bas-designer och dessa används sedan för att implementera minsta avståndsfunktionen. Denna minsta avståndet-funktion är sedan ämnad att implementeras som begränsningar för hur nära komponenter kan placeras varandra i ett program för design av mönsterkort vars syfte är att möjliggöra folk utan kunskaper inom mönsterkortsdesign att ändå kunna realisera sina designidéer.
|
2 |
Exploring feasibility of reinforcement learning flight route planning / Undersökning av använding av förstärkningsinlärning för flyruttsplanneringWickman, Axel January 2021 (has links)
This thesis explores and compares traditional and reinforcement learning (RL) methods of performing 2D flight path planning in 3D space. A wide overview of natural, classic, and learning approaches to planning s done in conjunction with a review of some general recurring problems and tradeoffs that appear within planning. This general background then serves as a basis for motivating different possible solutions for this specific problem. These solutions are implemented, together with a testbed inform of a parallelizable simulation environment. This environment makes use of random world generation and physics combined with an aerodynamical model. An A* planner, a local RL planner, and a global RL planner are developed and compared against each other in terms of performance, speed, and general behavior. An autopilot model is also trained and used both to measure flight feasibility and to constrain the planners to followable paths. All planners were partially successful, with the global planner exhibiting the highest overall performance. The RL planners were also found to be more reliable in terms of both speed and followability because of their ability to leave difficult decisions to the autopilot. From this it is concluded that machine learning in general, and reinforcement learning in particular, is a promising future avenue for solving the problem of flight route planning in dangerous environments.
|
Page generated in 0.0405 seconds