Robotics and Autonomous Systems
Objectives
- . Understanding
- Perception for Autonomous Systems basic concepts
- Main Robot Sensors and their characteristics
- Robot 2D and 3D Vision
- Perception using and fusing diverse modalities
- Active vs Passive Perception
- Probabilistic Robot Localisation methods
- . Able to Do
- Addressing new problems and implementing perception and localisation strategies in the domain of robotized heterogeneous autonomous systems
- Increase the capacity to practically implement these systems in robotized autonomous platforms
- Apply creativity and innovation
- Non-Technical Competences
- Develop synthesis critical thinking
- Team working and increasing oral and writing communication skills
- Improve time keeping and compliance with meeting deadlines
General characterization
Code
10991
Credits
6.0
Responsible teacher
José António Barata de Oliveira, Luís Manuel Camarinha de Matos
Hours
Weekly - 4
Total - 56
Teaching language
Português
Prerequisites
- Programming Skills
Bibliography
- Thrun, S, M., Burgard, Fox D (2006). Probabilistic Robotics. MIT Press,668
- Siegwart, R., Nourbakhsh, I. R., & Scaramuzza, D. (2011). Introduction to autonomous mobile robots. MIT press.
- Chatterjee, A., Rakshit, A., & Singh, N. N. (2012). Vision based autonomous robot navigation: algorithms and implementations (Vol. 455). Springer.
- Ferreira, J. F., & Dias, J. M. (2014). Probabilistic approaches to robotic perception. Springer International Publishing.
- Kaehler, A., & Bradski, G. (2016). Learning OpenCV 3: computer vision in C++ with the OpenCV library. " O''Reilly Media, Inc.".
Teaching method
Theoretical-practical classes (TP) are directed so that students, through their active participation, understand each of the topics listed in the learning objectives.
In laboratory classes (PL) students focus on the experimentation of the concepts exposed in theoretical-practical classes in order to know how to do.
For each practical work:
- Presentation of the work,
- tutorial on the technology / tools to use,
- discussion of the work method,
- realization of the work by the students accompanied by teachers, and
- preparation of report.
Evaluation Components
- 2 Mini-Tests
- 3 Practical Works
Evaluation Rules
- Theoretical Mark = (Mini-Test 1 + Mini-Test 2) / 2
- Theoretical Mark >= 9.5
- Each Practical Work >= 9.5
- Practical Mark = TP1 * Weight1 + TP2 * Weight2 + TP3*Peso3 ; Weights to be announced at the beginning of UC
- Final Mark = Practical Mark * 0.7 + Theoretical Mark * 0.3
Evaluation method
Evaluation Components
- 2 Mini-Tests
- 3 Practical Works
Evaluation Rules
- Theoretical Mark = (Mini-Test 1 + Mini-Test 2) / 2
- Theoretical Mark >= 9.5
- Each Practical Work >= 9.5
- Practical Mark = TP1 * Weight1 + TP2 * Weight2 + TP3*Peso3 ; Weights to be announced at the beginning of UC
- Final Mark = Practical Mark * 0.7 + Theoretical Mark * 0.3
Subject matter
1. INTRODUCTION
o Perception for Robotics: Measurement vs. Perception ( 2D, 3D, Sound, etc)
2. VISION BASED SENSORS:
o Monocular Camera, Binocular Camera
o Omnidirectional, Mutlispectral Imagery
3. THREE-DIMENSIONAL DATA:
o Streoscopic Pair, Structured Light
o LiDAR, SONAR, Time of Flight
4. THREE-DIMENSIONAL OCCUPANCY REPRESENTATION:
o Point clouds, Voxel Grids, Octrees
5. THREE-DIMENSIONAL REGISTRATION
o Iterative Closest Point
o Scan Matching
6. THREE-DIMENSIONAL FILTERING THREE-DIMENSIONAL SEGMENTATION
o Sample Consensus
o Euclidean & Conditional Clustering
o KdTrees"
7. MULTIMODAL PERCEPTION AND DATA FUSION
8. MULTIMODAL BASED OBJECT TRACKING
9. ACTIVE PERCEPTION
10. MOBILE ROBOTS LOCALISATION AND MAPPING
o The challenge of Localisation
o Dead Reckoning
o Basic Localisation Sensors"
o Global Localisation
o Map based Localisation
o SLAM techniques