Artificial Intelligence and Autonomous Systems (M2-MARS / ASI) - WEUMAIA0

Goal(s)

The purpose of this course is to give an overview of the main techniques of artificial intelligence and a thorough understanding of how to apply them to the robotics and autonomous systems.

At the end of this course, students will be able to:

• Describe and identify the essential blocks of an autonomous system
• Develop autonomous navigation strategies by making assumptions about the environment and integrating the dynamic model of the mobile agent/robot
• Analyze and model the interconnections of multi-agent systems, developing basic strategies for multi-robot coordination.
• Explain and describe the basic methods and fundamental statistical/mathematical tools used in supervised and unsupervised learning
• Compare the characteristics, capabilities and limitations of these different approaches
• Apply these methods to real datasets in the fields of signal and image processing

Content(s)

*Autonomous Systems : Introduction to AI FOR AUTONOMOUS SYSTEMS (30 hours, including courses and lab work)

This part of the course aims at studying the essential blocks of autonomous systems with a focus on planning and navigation techniques. The basic tools to deal with multi-agent systems will be analyzed and several applications for multi-robot systems will be studied.

– Introduction to autonomous systems
– Autonomous Navigation for Mobile Robots
o Representation of the environment
o Path planning
o Artificial potential fields for navigation
– Cooperative Multi-Robots Systems
o Introduction to Graph theory
o Consensus algorithms
o Flocking
o Cooperative Coverage
o Cooperative Exploration

*Artificial Intelligence: Introduction to MACHINE and STATISTICAL LEARNING (30 hours, including courses and lab work)

MACHINE LEARNING and Statistical LEARNING PRINCIPLES (ML)
This part of the course aims at providing the students with the basic methods and statistical/mathematical tools to understand the most classical approaches used in both supervised or unsupervised learning, as well as understanding their capabilities, characteristics and limitations.

ML-1: "Introduction to Machine Learning
- motivations
- basic challenges: classification/regression, Supervised/unsupervised, cost functions, training error/testing
- examples of approaches WITH and WITHOUT models
- Notion of complexity: variance bias trade-off (over or under learning), scourge of the dimension

ML-2: Generative Approaches :
- Discriminant analysis (linear/quadratic, regularization, size reduction)
- Naive Bayesian (parametric and non-parametric
- examples given MNIST, ...

ML-3: Discriminatory approaches :
- generalized linear statistical models for regression and classification (logistic regression)
- ridge regulation, lasso (parsimony bet)
- Complexity and validation (cross-validation) of models
- Data transformation (kernel move)
- Examples given MNIST, ...

ML-4: "DATA REPRESENTATION" (Unsupervised).
- BCP, standardization, small visualization

ML-5. "(Convolutional) Neural Nets."
- Notion of neuronal learning, gradient stoch. GRADIENT, NN
- Deep, convolutional aspects... motivations and examples (scikit-learn, kheras/tensorflow,...)

ML-6. Random trees and forests
- decision tree, random forests

ML-7. "CLUSTERING
- Kmeans, Gaussian mixture model, EM algorithm,
- Ex MNIST data and comparison with CNN
- hierarchical patterns

Probabilities and statistics, main distributions, probability density functions, matrix calculation, basic optimization (Newton's methods, Gradient methods...), basic control theory

The math reminders will be done in general (the prerequisites having been seen in 1A and 2A), but very quickly.