AI Domains of Interest for Intelligent Robotics

Domain Brief Definition
Semantic and Cognitive Robots Semantic addresses the processing of commands in ordinary spoken language, following signs, and interpreting sentences in either written or spoken form. Cognitive deals with mental activities such as thinking, understanding, learning, and reasoning undertaken in decision making processes using machine perception, planning, learning, and reasoning
Autonomous Navigation Guiding motion based on fusion of sensory data and intelligent data processing. Integrates motion control, communication (Ethernet, Wi-Fi, 4G/LTE) and sensor (Lasers, Depth Cameras, Ultrasound Sensors, IR sensors). Module enables to construct autonomously 3D maps, localize, path tracking and integrate obstacle avoidance procedures
System Control Based on Primitives Task and motion primitives are an effective way of representing complex tasks, to reduce the search for control algorithms, make model identification or learning in high dimensional systems feasible, and facilitate model online adaptation. The task-motion primitive paradigm is particularly well suited for object manipulation tasks such as grasping of unknown objects with a robot hand.
Machine (Reinforcement) Learning A paradigm of learning by trial-and-error, solely from rewards or punishments that trains algorithms for providing an indication if a correct decision or wrong decision was made.  With enough iterations a reinforcement learning system will eventually be able to predict the correct outcome. A reinforcement learning algorithm, or agent, learns by interacting with its environment. The agent receives rewards by performing correctly and penalties for performing incorrectly. The agent learns without intervention from a human by maximizing its reward and minimizing its penalty to maximize performance based on feedback from the environment.
Object Detection & Recognition Static and dynamic object recognition based on using image classification, image segmentation and object labelling.
Face Detection & Recognition Uses features identification for classification and recognition.
Facial and Emotional Expression Detection Extraction of emotional state based on voice and face features recognition.
Voice Communication Expressing natural voice in terms of computer language.
Human-Machine Interface Graphical and physical device interfaces.
Grasping & Manipulation Object modeling to obtain contact points for handling objects with robot hands and allow their manipulation.
Obstacle Detection & Avoidance Identification of obstacles without classification and avoidance of collision.
Power Recharging Provide autonomous robot motion to recharging stations.
User Data Storage  and Retrieval Classify information and allow random access retrieval.
Use of Database on Cloud High-speed broad bandwidth link to mass storage.
iOS-based API Provide interfaces for user software and hardware applications.