The literature section is classified into the following subsections. The first subsection addresses the modelling aspects of the intelligent robotic systems. The second subsection deals with the control aspects of the robotic mechanisms. The third and the fourth subsections respectively emphasize the fuzzy and other artificial intelligence (AI) based applications in a robotic environment. The fifth subsection narrates the medical and surgical-based applications of the robots. The sixth subsection discusses swarm robots. The seventh subsection deliberates on the communication between robots and between robot and operator as well as robotic systems dealing with navigation-based methods. The final subsection takes into account some miscellaneous applications such as space exploration, robot vision system and robotic-assisted systems for the welfare of humans.

Modelling of Intelligent Robotic Systems

Valvanis et al. [23] presented the hardware and software design for the three modelling levels of intelligent modelling systems. All the three hierarchical levels were modelled differently, and microprocessor and discrete logic-based techniques were proposed for the hardware. Later, an algorithm in the assembly and organization level of robots with different constraints was utilized for the optimization of planning procedure for each system in order to make them capable for doing the requested jobs [24]. A probabilistic approach was presented in Reference [25], in which the hierarchical level of performing tasks was employed. In order to mathematically interpret the system functions, probability and entropy functions were used. Valvanis et al. [26] proposed a general-purpose method for modelling robotic assemblies and intelligent robotic systems, which considered both the pre-defined and fuzzy commands.

Yang et al. [27] decomposed the uncertain robotic model into repetitive and non- repetitive. The Lyapunov model was used for developing iterative control methods for robots having structured certainty and uncertainty. Martinoli et al. [11] presented discrete-time methodology for incremental modelling at the microscopic and macroscopic levels for manipulation using the swarm autonomous robotic systems and reviewed the then existing methodologies by proposing a few changes. Hu et al. [28] designed the team-formation mechanism of the multi-robot system and used it for modelling purposes on the basis of discrete event system specification (DEV) formalism and proved the process of simulation based on the physical mapping of DEVs. Stojanov et al. [29] discussed the overview of the research done in robotics and how curiosity was associated with it through a one-dimensional and reductionist approach. The modelling of curiosity in context of cognitive architecture was also presented. Wurm et al. [30] presented an approach consisting of 3D modelling algorithms using a probabilistic estimation and also reviewed all the models existing then for 3D modelling. The results proved that the probabilistic approach w'as able to model the robotic systems. Collins et al. [31] illustrated basic principles and algorithms developed for the realization of intelligent robotics. The book also dealt wdth stability and validations for the kinematics and dynamics of the robot.

Control Aspects of Intelligent Robotic Systems

Kazerooni et al. [32] described the dynamics of extender robotic systems which are worn by human beings for performing various tasks. Humans exchange both power and signal w ith the extenders, and an arbitrary relationship has been derived between the human and load force. Dubowsky et al. [33] discussed the dynamics and control issues for the free-flying and free-floating-based robots used for different tasks and presented three methods for the planning and controlling of these robots. Sugano et al. [34] studied the stability of the manipulators for vehicles, assuming that they have relation with the vehicle speed, and presented the concept of stability and region of stable operation using ZMP criteria. Tarn et al. [35] applied event-based planning and showed that the representations in path-based and time-based feedback systems have non-linear nature in order to obtain stability. The results w'ere also verified using experiments. Vemuri et al. [36] investigated the problems faced in the fault detection of rigid-link robotic systems, including all the uncertainties, and established the robustness and stability. The use of neural networks had been applied in the robust fault detection and stability of system. Silva [37] reviewed the techniques used for integrating intelligence in the control systems of robots. Also, many concepts related to approximations were explored and presented a generalized model for knowledge- based decision-making. Fukuda et al. [38] studied some of the intelligent techniques for the control of robotic systems and introduced a network of robotic systems and discussed major schemes for the different hierarchical levels of the robotic systems. Tarn et al. [39] presented new control and planning for intelligent robotic systems significantly related to their motion by the combination of closed event-based plan with the non-linear feedback and also presented stability criteria. Jacak [40] presented a method for realization of intelligent robotic systems, including some of the important tools, and described the necessary actions for planning, coordination and control of robots. Choi et al. [41] proposed an adaptive iterative learning control (AILC) scheme for the stability of the robotic system including a balanced conventional adaptive control and iterative control and complemented the disadvantage of each method. Behai et al. [42] described the use of Lyapunov-based systems for the design of a non-linear controller which can help to solve the problems faced by the robot in manipulating the environments. Antonelli et al. [43] proposed a controller based on prioritized task kinematics to provide error stability and convergence. For prioritizing the data and its stability, Lyapunov-based stability was used base on the kinematics algorithm. Short et al. [44] described a method to develop design philosophy and reference architecture for advanced robotic architecture and also described a prototype using the same method which was used for describing the controller architecture. Krishnan et al. [19] reviewed articles on assistive devices which focused on self-transfer and mobility systems. Also, they covered the advances made in the above-mentioned field along with its limitations. Na et al. [45] studied the parameter estimation and controlling of the non-linear robots by minimizing the errors. The equation for parameter estimation was based on the auxiliary filtered variables and investigated robustness against disturbances.

< Prev   CONTENTS   Source   Next >