• Home
  • Mohammad Ali Zamani
Mohammad Ali Zamani

Mohammad Ali Zamani
Hamburger Informatik Technologie-Center (HITeC) · R&D

PhD fellow

About

27
Publications
4,823
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
238
Citations
Citations since 2017
11 Research Items
154 Citations
2017201820192020202120222023010203040
2017201820192020202120222023010203040
2017201820192020202120222023010203040
2017201820192020202120222023010203040
Introduction
(Homepage: zamani.github.io) a machine-learning scientist with a passion for finding scalable solutions based on deep learning for industrial problems. My research focus has been on deep reinforcement learning including explainability and its applications in robotics, dialogue management systems, and planning. I am also interested in AutoML as a path to reach a scalable artificial intelligence solution.
Education
May 2016 - May 2020
University of Hamburg
Field of study
  • Computer Science
September 2012 - August 2015
Ozyegin University
Field of study
  • Computer Science
September 2004 - July 2009
University of Tehran
Field of study
  • Electrical Engineering (Major: Control Engineering)

Publications

Publications (27)
Chapter
Full-text available
Deep Reinforcement Learning (DRL) has become successful across various robotic applications. However, DRL methods are not sample-efficient and require long learning times. We present an approach for online continuous deep reinforcement learning for a reach-to-grasp task in a mixed-reality environment: A human places targets for the robot in a physi...
Conference Paper
Deep Reinforcement Learning (DRL) has become successful across various robotic applications. However, DRL methods are not sample-efficient and require long learning times. We present an approach for online continuous deep reinforcement learning for a reach-to-grasp task in a mixed-reality environment: A human places targets for the robot in a physi...
Conference Paper
Full-text available
Previous work on emotion recognition demonstrated a synergistic effect of combining several modalities such as auditory, visual, and transcribed text to estimate the affective state of a speaker. Among these, the linguistic modality is crucial for the evaluation of an expressed emotion. However, manually transcribed spoken text cannot be given as i...
Preprint
Full-text available
Previous work on emotion recognition demonstrated a synergistic effect of combining several modalities such as auditory, visual, and transcribed text to estimate the affective state of a speaker. Among these, the linguistic modality is crucial for the evaluation of an expressed emotion. However, manually transcribed spoken text cannot be given as i...
Article
Full-text available
Spoken language is one of the most efficientways to instruct robots about performing domestic tasks. However, the state of the environment has to be considered to plan and execute actions successfully. We propose a system that learns to recognise the user’s intention and map it to a goal. A reinforcement learning (RL) system then generates a sequen...
Conference Paper
Full-text available
We present a neural end-to-end learning approach for a reach-for-grasp task on an industrial UR5 arm. Our approach combines the generation of suitable training samples by classical inverse kinematics (IK) solvers in a simulation environment in conjunction with real images taken from the grasping setup. Samples are generated in a safe and reliable w...
Conference Paper
Full-text available
Speech emotion recognition (SER) is an important aspect of effective human-robot collaboration and received a lot of attention from the research community. For example, many neural network-based architectures were proposed recently and pushed the performance to a new level. However, the applicability of such neural SER models trained only on in-dom...
Conference Paper
Full-text available
Robotic motor policies can, in theory, be learned via deep continuous reinforcement learning. In practice, however, collecting the enormous amount of required training samples in realistic time, surpasses the possibilities of many robotic platforms. To address this problem, we propose a novel method for accelerating the learning process by task sim...
Conference Paper
Full-text available
Acoustically expressed emotions can make communication with a robot more efficient. Detecting emotions like anger could provide a clue for the robot indicating unsafe/undesired situations. Recently, several deep neural network-based models have been proposed which establish new state-of-the-art results in affective state evaluation. These models ty...
Conference Paper
Full-text available
Spoken language can be an efficient way to warn robots about threats. Guidance and warnings from a human can be used to inform and modulate a robot's actions. An open research question is how the instructions and warnings can be integrated in the planning of the robot to improve safety. Our goal is to train a Deep Reinforcement Learning (DRL) agent...
Conference Paper
Full-text available
Programming robots for a safe interaction with humans is extremely complex especially in collaborative tasks. One reason is the unpredictable behaviour of humans that may have an intention which is not clear to the robot. We present a novel architecture for a safe human-robot collaboration scenario in a shared tabletop workspace based on intuitive...
Chapter
Full-text available
This chapter expresses three cognitive genres: descriptive genre, normative/prescriptive genre, and know-how genre. The descriptive genre introduces and discusses on the following disciplines: the core concepts of complexity, complex adaptive system (CAS) of systems (CASoS), the application domains of human-automation interaction (HAI) and adaptive...
Conference Paper
We earlier introduced a novel framework for realization of Adaptive Autonomy (AA) in human-automation interaction (HAI). This study presents an expert system for realization of AA, using Support Vector Machine (SVM), referred to as Adaptive Autonomy Support Vector Machine Expert System (AASVMES). The proposed system prescribes proper Levels of Auto...
Article
The complexity of humans and automation interaction in Smart Grid, as the future of power system, calls for embedding a level of intelligence in the system. Adaptive Autonomy (AA) theory is employed to manage high level complexity of the Human-Automation Interaction (HAI) system. The fuzzy expert system with Gradient Descent Algorithm (AAFGDES) is...
Article
while Smart Grid (SG) expectations call for automation of power distribution systems, disregarding human factors in the Distribution Automation (DA) systems can make it more problematic than beneficial. Thus, human factors and automation systems should be considered simultaneously. However, the interaction of humans and automation systems involves...
Article
Full-text available
this paper presents a novel expert system referred as Petri Net equipped Tricotyledon Theory of System Design (PN- T3SD) to select a proper IT infrastructure for smart grid. The proposed PN-T3SD is a policy-driven decision making method - combining the ideas from Petri nets and Wymorian T3SD- that can change the result of decision making according...
Chapter
Control of an unknown nonlinear time-varying plant has always been a great concern for control specialists, thus an appealing subject in this discipline. Many efforts have been dedicated to explore the various aspects of this problem. This research has led into introducing many new fields and methods. These methods can be categorized into two gener...
Article
Interaction of human and computer agents should be harmonized by adapting the automation level of the IT systems to maintain a high performance for the system in changing environmental conditions. This research presents an expert system for the realization of adaptive autonomy (AA), using Petri Net (PN), referred to as AAPNES, based on practical li...
Article
Full-text available
This paper presents an expert system for selecting a proper IT infrastructure for smart grid, based on a novel fuzzy approach to Tricotyledon Theory of System Design (Fuzzy-T3SD). The proposed Fuzzy-T3SD is a policy-driven decision making method that can change the result of the decision makings according to the utility's policy. The Fuzzy-T3SD is...
Conference Paper
Smart grid expectations objectify the need for optimizing power distribution systems greater than ever. Distribution Automation (DA) is an integral part of the SG solution; however, disregarding human factors in the DA systems can make it more problematic than beneficial. As a consequence, Human-Automation Interaction (HAI) theories can be employed...
Conference Paper
Full-text available
Power grid cyber security is turning into a vital concern, while we are moving from the traditional power grid toward modern Smart Grid (SG). To achieve the smart grid objectives, development of Information Technology (IT) infrastructure and computer based automation is necessary. This development makes the smart grid more prone to the cyber attack...
Conference Paper
While the smart grid dream is likely to be revisited, we are witnessing a convergence of Information Technology (IT) with power system engineering. This convergence should be managed tactfully, to obtain a consistent architecture for the smart grid. Due to the complex nature of the both IT and power systems, a higher degree of complexity emerges th...
Conference Paper
A safe and secure world needs to effective collaboration between humans and intelligent automation systems, as research on human-automation interaction (HAI) can establish more trust in automation. This paper is intended to investigate the relationship between the human-automation systems and the factors which shape their performance, referred to a...
Conference Paper
We have introduced a novel framework for realization of Adaptive Autonomy (AA) in human-automation interaction (HAI) systems, as well as several expert system realizations of that. This study presents an expert system for realization of AA, using logistic regression (LR), referred to as Adaptive Autonomy Logistic Regression Expert System (AALRES)....
Article
Full-text available
Earlier we introduced a novel framework for implementation of Adaptive Autonomy (AA). This study presents an expert system realization of the AA framework, referred to as Adaptive Autonomy Expert System (AAES). The proposed AAES is based on the extracted rules from the Expert’s Judgment on proper Levels of Automation (LOA) for various environmental...
Conference Paper
Intelligent control and automation is associated with expert systems; especially, when it needs to human expertise. Earlier we introduced a framework for implementation of adaptive autonomy (AA) in human-automation interaction systems, followed by a data-fusion-equipped expert system to realize that. This paper uses fuzzy sets concept to realize th...

Network

Cited By

Projects

Projects (2)
Archived project
CONVERGE project aims to improve the efficacy of a robot skill generation framework by allowing the human and robot learn simultaneously, and work together as a team.
Project
SECURE is a new Marie Skłodowska-Curie Action funded by the European Commission. It's aim is to train roboticists and research fellows on the cognitive and interaction level of robot safety. These fellows should then be able to cope with the new challenges for safety that come with the increased complexity in human work and living spaces. They also need to be familiar with safety concepts and solutions for a multitude of robotic platforms. Therefore, the SECURE network aims to train fellows on innovative scientific and technological requirements for safe human-robot interaction and will employ several of the currently best robot platforms in Europe. The fellows are trained at six partner institutions in Europe and are supported by another five associated partners, ranging from large-scale international industrial partners to small enterprises, thus providing an optimal training environment for young researchers. Official website: http://secure-robots.eu/