Sökning: "kandidatuppsats datavetenskap"

Visar resultat 1 - 5 av 32 uppsatser innehållade orden kandidatuppsats datavetenskap.

  1. 1. Reinforcement Learning for Multi-Agent Strategy Synthesis Using Higher-Order Knowledge

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Gustav Forsell; Shamoun Gergi; [2023]
    Nyckelord :Higher Order Knowledge; Imperfect Information; Reinforcement Learning; Deep Q- networks; Knowledge Representation; Pursuit Evasion Games;

    Sammanfattning : Imagine for a moment we are living in the distant future where autonomous robots are patrollingthe streets as police officers. Two such robots are chasing a robber through the city streets. Fearingthe thief might listen in to any potential transmission, both robots remain radio silent and are thuslimited to a strictly visual pursuit. LÄS MER

  3. 2. Near-Real Time Forest Fire Monitoring System From an UAV

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :August Näsman; Daniel Nedlich; [2023]
    Nyckelord :;

    Sammanfattning : The purpose of this thesis is to implement a payload system on a drone to help fire towers in near-realtime survey forests for wildfires. The payload system should be able to communicate with a groundstation through a mobile network and the survey should be tagged with relevant metadata. LÄS MER

  4. 3. Multi-Robot Motion Planning Under High-Level Task Specifications

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Axel Abrahamsson; Lukas Granqvist; [2023]
    Nyckelord :;

    Sammanfattning : This bachelor thesis explores the use of Signal Temporal Logic (STL) and ControlBarrier Functions (CBFs) to address the challenges associated with multi-robot motionplanning under high-level task specifications. STL is a formalism used to specify temporalproperties of signals, while CBFs are used to enforce safety constraints. LÄS MER

  5. 4. Predicting Asset Indexes for Safe and Profitable Portfolio Allocation

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Ludwig Fredriksson; Nikola Tomic; [2023]
    Nyckelord :;

    Sammanfattning : Many investors face the complicated task of allocating and forecasting asset indexes in asafe and profitable manner. The primary objective of this bachelor thesis is to introduce asafe and risk-adjustment portfolio allocation, consisting of four Swedish listed. LÄS MER

  7. 5. Implementing a Network Optimized Federated Learning Method From the Ground up

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Gustav Källander; Henning Norén; [2023]
    Nyckelord :;

    Sammanfattning : This bachelor thesis presents the implementation ofa simple fully connected neural network (FCNN) and federatedneural network with stochastic quantization from scratch andcompares their performance. Federated learning enables multipleparties to contribute to a machine learning model withoutsharing their sensitive data. LÄS MER