Laxmisha Rai

Multi-thread based synchronization of locomotion control in snake robots

In this paper, we present an approach to control the locomotion of snake robot with concurrent programming model constructed using threads and semaphores. The multi-thread based concurrent programming model adds the flexibility to design and synchronize the movement of snake robots as compared with microcontroller and mechanical based approaches. We have designed a physical snake robot using LEGO sensors and actuator blocks and the wave motion of the snake robot is generated by multi-thread based concurrent programming under RT-Linux. The different robot movements in a desired direction along with different types of snake movements are achieved using angle sensors.

Knowledge-Based Integration Between Virtual and Physical Prototyping for Identifying Behavioral Constraints of Embedded Real-Time Systems

Rapid prototyping methods are in need of autonomous decision making and analysis during the product development stages so that the ldquotime-to-marketrdquo can be reduced faster than traditional product development methodologies. Therefore, new methods of prototyping are inevitably essential. This paper proposes an approach to utilize the benefits of virtual prototyping (VP) and physical prototyping (PP) methodologies by integrating them into knowledge-based systems (KBSs) by providing seamless connection. This approach is termed autonomous integrated prototyping. The main contribution of this paper is the development of an intelligent system architecture to facilitate and guide the product development autonomously and simultaneously in both VP and PP environments. The seamless connection between VP and PP, along with KBSs, enables the exploration of new behaviors of developing systems and analyzing different behaviors. The architecture is applicable to embedded real-time systems (ERTSs), sensor applications, robotics, and ubiquitous applications where system interaction with the external environment is necessary.

RT-Linux based hard real-time software architecture for unmanned autonomous helicopters

The UAV (unmanned aerial vehicle) performs various kinds of missions while performing autonomous flight navigation. In order to realize all functionalities of the UAV, the software part becomes a very complex hard real-time system because some hard real-time tasks, soft real-time tasks and non-real-time tasks are concurrently working together under a RTOS. In this paper, a hierarchical architecture for unmanned autonomous helicopter system is proposed that guarantees the real-time performance of hard real-time tasks and the re-configurability of soft realtime or non-real-time tasks under the RT-Linux. This software architecture has four layers: hardware, execution, service agent and remote user interface layer according to the reactiveness level for external events. In addition, the layered separation of concurrent tasks makes different kinds of mission reconfiguration possible in the system. An unmanned autonomous helicopter system was implemented using Kyosho RC Helicopter to test and evaluate the performance of the proposed systems.

Rule-based modular software and hardware architecture for multi-shaped robots using real-time dynamic behavior identification and selection

Multi-shaped robots can change their shape to navigate different terrains and thereby accomplish complex tasks. Such robots are more useful in rough terrains, pipeline inspections and in the removal of obstacles. The design of such multi-shaped robots requires dynamic shape changes in its hardware as well as reconfigurable software architecture. Modular robots are preferred under such conditions, in that they can change their behavior dynamically. This paper proposes new modular software and hardware architecture for multi-shaped robots using real-time dynamic behavior identification and selection. It is a layered architecture with reusable and reconfigurable modules, which can be embedded in an expert system as both hardware and software modules and demonstrated with snake robot or physically reconfigured four-legged robots. The intelligent dynamic selection and synchronization of selected behaviors enable mobile robots to perform many tasks under complex situations. The architecture proposed can be applied to shape-reconfigurable robots, for the dynamic selection of behavior during reconfiguration, where hardware and software modules can be reused during reconfiguration.

Remote Controlled Group Behavior for Widely Spreaded and Cooperative Mobile Robots in Wireless Sensor Network Environment

In this paper, we present the experiments with robots in a wireless sensor network environment to support intelligent generation of group behaviors. We propose real-time software architecture in a wireless sensor network (WSN) environment for practical applications. The architecture is a layered architecture including decision making, knowledge processing, execution, and communication and sensor/actuator layers. The proposed architecture is tested for the multi-robot environment, where the robots are expected to exhibit group behaviors. The rules are written in CLIPS expert system tool to perform intelligent behavior generation and dynamic reasoning so as to make the behaviors more realistic.

Non-Deterministic Behavior Modeling Framework for Embedded Real-Time Systems Operating in Uncertain Environments

While complex embedded real-time systems (ERTS) such as robots in operation, there is a possibility that unstructured and unrelated data may be gathered over a period of time through sensors and may result in unexpected behaviors or catastrophes. Without proper modeling of non-deterministic behaviors, implementing highly expected results to handle complex situations is expensive to the designers and may result in numerous programming challenges. For analyzing such situations, a stable and general modeling framework to support the designers for rapid analysis of the system behavior is needed. This paper proposes a generic behavioral modeling framework for embedded real-time systems in uncertain environments based on the few empirical studies. The key contribution of the paper is to develop a framework which can be applied to many ERTS applications, where the system behaviors can be predicted exactly during system in operation. Moreover, the architecture gives overall flexibility to apply all possible behaviors in different situations dynamically. The behaviors are generated by applying various facts and rules which are mapped to tasks. As the limited number of tasks may generate unlimited number of rules and thus unlimited number of behaviors, the modeling architecture provides a best possible way to optimize the necessary behaviors and completely discard the less useful behaviors.

Intelligent embedded real-time software architecture for dynamic skill selection and identification in multi-shaped robots

This paper presents an intelligent embedded, modular software and hardware architecture for multi-shaped robots using real-time dynamic skill identification and selection. It is a layered architecture with reusable and reconfigurable modules, which can embed in an expert system as both hardware and software modules and demonstrated with snake robot and physically reconfigured four-legged robot as examples. The intelligent dynamic selection and synchronization of selected behaviors enable the mobile robot to perform many tasks in complex situations. The architecture proposed is applicable to multi-shaped robots, for dynamic selection of behaviors during reconfiguration, where the hardware and software modules can be reused during reconfiguration. Related videos of these robots can be viewed at: http://rtlab.knu.ac.kr/robots.htm

Behavior Modelling of Embedded Real-Time Systems Operating in Uncertain Environments

Decisions made at uncertain environments by the embedded real-time systems (ERTS) have a long-term effect on the surroundings and users usually expect the system to behave intelligently. Implementation of expected behaviors and identifying nondeterministic behaviors of such systems operating in uncertain environments is a challenging issue for the embedded system designers. To cope with multiple tasks running in the same or several sub-systems and to manage the complexity of control as well as to improve the performance of the system, there is a need for stable modeling framework. This paper proposes a framework which can be applied to many ERTS applications, where the system behaviors can be predicted exactly during system in operation. Moreover, the architecture gives overall flexibility to apply all expected behaviors in different situations dynamically based on sensor responses.