0x010088F6312065A70A44B630A6848498EAD80093CFEAFE161DC94D83C7C5C9A5A748B8

<h1>1. Course Description</h1> In this part, course description of the Swarm Intelligence course in autumn 2022, will be introduced. <h2>1.1.  Course Objectives</h2> This course is designed to present an overview of Swarm Intelligence (SI) topic, including both behavioural swarm Intelligence and computational swarm intelligence, and applications of SI. The students will learn different swarm intelligence algorithms that are inspired by natural systems such as ant colonies, bird <a href="https://en.wikipedia.org/wiki/Flocking_%28behavior%29" target="_blank">flocking</a>, animal <a href="https://en.wikipedia.org/wiki/Herding" target="_blank">herding</a>, bacterial growth, fish schooling and microbial intelligence. The students will implement different swarm intelligence algorithms, visualize, and apply them to solve real problems such as optimization problems.  <h2>1.2. Course Content </h2>Course topics: There are four main topics: <ol><li>Agent-based modeling: Bottom-up modeling method. individual agents. System theory and complex systems. Multi-agent systems.</li><li>Behavioral swarm intelligence: Modeling flocking behavior. Boids model. Flocking behavior applications, such as agents queuing and homing.</li><li>Computational swarm intelligence (CSI): Optimization theory and multi-objective optimization. Particle swarm optimization (PSO) Ant colony optimization (ACO). Bees colony algorithm (BCO). Bats algorithm</li><li>Selected applications: Different selected application where the students can apply the swarm intelligence algorithms to solve real problems, such as:</li></ol>·      Multi-robot path planning·      Task scheduling.·      Etc. <h2>1.3.  Learning outcome</h2> Upon completion of the course, students will be expected to:Knowledge: <ul><li>Have knowledge of individual/intelligent agents for modeling of industrial, social and biological systems.</li><li>Have knowledge of modeling swarms/social agents in complex landscapes. </li><li>Have knowledge of swarm intelligence algorithms inspired by different natural systems.</li></ul>Skills: <ul><li>Have practice in programming virtual worlds in a game engine.</li><li>Have skills in using individual/intelligent agents to solve optimization problems in complex landscapes.</li><li>Have skills in developing simulation models based on swarms of intelligent agents.</li><li>Have skills in using swarm intelligence algorithms to solve real optimization problems.</li></ul>General competence: <ul><li>Have general knowledge about the subject's possibilities and limitations.</li><li>Have general knowledge of being able to analyze, disseminate and communicate the topic issues.</li><li>Have general knowledge about how intelligent agents can contribute to innovation processes.</li></ul><h2>1.4. Teaching method and activities</h2> ·     Lectures, discussion at group and class level, exercises, student presentations, mandatory assignments covering the whole course. The mandatory assignments are performed individually.·     Three  mandatory assignments must be passed for permission to enter the oral exam. <h2>1.5. Curriculum</h2> he course material will be taken from different books and white papers. Main course textbook:Andries P. Engelbrecht, Fundamentals of computational swarm intelligence, Wiley (2015), ISBN: 978-0-470-09191-3.Other relevant textbooks: <ul><li>Anand Nayyar, Dac-Nhuong Le, Nhu Gia Nguyen, Advances in Swarm Intelligence for Optimizing Problems in Computer Science, 1st Edition, (2018), Chapman and Hall/CRC, ISBN 9781138482517.</li><li>James Kenndey and Rusell C. Eberhart, Swarm Intelligence, Mkf (2001), ISBN: 978-1-55860-595-4.</li><li>Dan Simon. Evolutionary Optimization Algorithms: Biologically Inspired and Population-based Approaches to Computer Intelligence, Wiley, 2013 Ed.</li><li>Jun Sun, Choi-Hong Lai and Xiao-Jun Wu, Particle Swarm Optimization - Classical and Quantum perspectives, CRC Press (2019), ISBN 9780367381936</li><li>Harald Yndestad: Swarm Intelligence Tutotial</li><li>Harald Yndestad: Agents and Landscapes as Complex Systems</li></ul>Other reading materials and tutorials will be announced at the beginning of the course. <h2>1.6. Evaluation</h2>  Oral exam based on the obligatory assignments and course content. <h2>1.7. Course Plan</h2><h2> </h2> 1 (24/08/2022) Agent system modelling <ul><li> Introduction (ABM, AI, SI, AI-app)</li><li>System theory</li><li>Complex systems</li><li>Intelligent agents </li></ul> 2 (31/08/2022) Introduction to Unity<ul><li> Unity 3D course by Farhad</li></ul> 3 (07/09/2022) Behavioural swarm Intelligence <ul><li> Flocking behaviour</li><li>Boids model </li></ul> 4 (17/09/2022)<ul><li> Social agents /app.</li><li>Steering behaviour and rules</li><li>Mandatory Exercise I (28.09.2022)</li></ul> 5 (21/09/2022) Swarm Intelligence Algorithms <ul><li> Introduction to optimization </li><li>Single objective vs. multi-objective optimization</li></ul> 6 (28/09/2022)<ul><li> Particle swarm optimization algorithm (PSO)</li></ul> 7 (05/10/2022)<ul><li> PSO for path planning </li><li>Re-planning in a time frame when a path is not feasible (dynamic obstacles) </li><li>Mandatory Exercise II </li></ul> 8 (12/10/2022)<ul><li> Binary and Discrete PSO</li></ul> 9 (19/10/2022)<ul><li> Ant Colony Optimization Algorithm (ACO)</li></ul> 10 (26/10/2022)<ul><li> Bees Colony Algorithm (BCO)</li><li>Mandatory Exercise III </li></ul> 11 (02/11/2022)<ul><li> Bats algorithm  </li></ul> 12 (09/11/2022)   Selected Topics<ul><li> Multi Robot path planning and Task Allocation</li></ul> 13 (16/11/2022)<ul><li> Swarm robotics </li></ul> 14 (23/11/2022) Revision  <ul><li> Revisiting some selected topics </li></ul>   <h1>2. Course Evaluation</h1> In this part, course evaluation process of the Swarm Intelligence course in Autumn 2022, will be introduced. <h2>2.1. Procedure</h2> In order to perform the course evaluation, the students selected three students to form the reference group, to represent them in the course evaluation meetings. All the reference group members are international students, since this is a master course (international program), and most of the students are non-Norwegians.   The class group is 18 students. 17 are master students and one bachelor student (computer sciences, taken this course as an elective course).   The evaluation meetings are distributed among the semester time. We have a short discussion in the class weekly with all the students. These short meetings provide a good opportunity for students to tell their opinions early in the course, to help the teacher to improve and fix the problems early in the course. We have two meetings, directly after starting the course, and at the end of the semester to discuss the whole course (course materials, lectures, teaching methods, and the evaluation method).    <h2>2.2. Feedback from reference group</h2> Relevance: The students find the course relevant to their study and useful overall. Feedback: There were two meetings where we discussed all the issues. The students were satisfied with class, course materials, and teaching method. Some of the students found difficulties in the exercises and assignments due to the lack of programming skills. We tried to solve this issue by having a full day workshop on Unity3D and C#. Also, we hire a teaching assistant to help the students in the programming.  Teaching method: This course is divided to workshops (weekly workshops). We start the workshop with short discussion about the previous workshop, then introduction to the topic and the plan for the rest of the day. The lecture is almost an hour and the rest of the workshop we have practical exercises where the students must implement their own solutions. The classroom is the master lab (NMK), and the lectures are recorded and uploaded to the blackboard.  Assignments:  We have exercises for each workshop. We have three mandatory assignments in the course where the students get practices and preparation to develop their own solutions for different problems. These assignments, aiming to let the students earn the needed skills. The students have some freedom in the assignments, they can choose the implementation tool and in the third assignment; they have to define the problem and choose different algorithms to solve it.   <h2>2.3. Attendance</h2> The attendance was just good in general (~80%). In the exercise’s sections, not many were attending. The teacher and the assistant were available in the exercise’s sections. Some the students (5 students) didn’t arrive Ålesund on time (beginning of the course), three of them arrived later (October). <h2>2.4. Exam results</h2> The final exam is an individual oral exam (30 min). The grades are shown below. All the students must pass the mandatory assignments to get the final exam. 11 students submitted and successfully completed the mandatory work.  The grades were very good and distributed as shown down:   Students Grade 5 A 4 B 2 C 0 D 0 E 0 F <h1>3. Conclusion </h1> The course was successful in achieving the learning outcomes. The results are very good, and the students developed their skills in the field. I have found that challenging the students to implement the algorithms themselves, work on practical problems and compare solutions, is very effective method in achieving the learning outcomes.  I think the course materials and the evaluation method are very good. the students improve their skills massively during the course. Using assistants (professionals in Unity3D and programming) help the students so much in solving the programming and technical issues. But the assistants lack pedagogical experience, and this must be improved.   In this semester, we have a problem that some of students got their acceptance in the program late and they couldn’t attend the course in the first weeks, and then they attended it digitally and they were not well prepared. Also, some students lack the needed programming skills, this can be solved by providing templates and good student assistants. I have asked the reference group many times to write the their report in teams and deliver it, but they didn’t do it.     <h1> </h1>

NTNU:TIA:REST

Institutt for IKT og realfag

Department of ICT and Natural Sciences

Ingen hjelpemidler tillatt.

Engelsk

Obligatoriske arbeidskrav

Swarm intelligence

Tekniske fag

Master i simulering og visualisering

Bachelor i ingeniørfag, data

Web/Lists(guid'8b698a2d-6ad0-4f7d-b3be-219cc1979f02')/Items(19483)/FieldValuesAsText

Web/Lists(guid'8b698a2d-6ad0-4f7d-b3be-219cc1979f02')/Items(13942)/FieldValuesAsText

Swarm intelligence - IE501714