Intelligent Embedded Systems
Abstract: Embedded Systems are a crucial technology for competitiveness. The vision of pervasive computing is that objects, buildings and environments may be endowed with software intelligence to improve human interactions both with the individual objects and with the system as a whole. Many intelligent embedded systems move rapidly within a physical environment. While the best complete algorithms are doubly exponential, probabilistic algorithms have emerged that have very good practical performance, and probabilistic guarantees of convergence.
Embedded Systems
'Intelligence' takes account of autonomous reasoning and acting in a co-operative manner. 'Ambient Intelligence' refers to an exciting new paradigm in information technology, "in which people are empowered through a digital environment that is aware of their presence and context and is sensitive, adaptive and responsive to their needs, habits, gestures and emotions." This applies not only for people-centred tasks, which, of course, seems the most exciting, science-fiction-type, aspect, but also for purely technical solutions like smart sensors, actuators and control systems, especially in safety related applications.
Heterogeneity (of environment, applications, protocols, etc.), autonomy (self-awareness, self healing, self-organizing, etc.), nomadic mobility (ad hoc, unreliable, heterogeneous, etc.) and scale-less (number of users, geography, structure, etc.) are the new emerging embedded systems challenges.
Used in everything from consumer electronics to industrial equipment, embedded systems —small, specialized computer systems stored on a single microprocessor — are playing a major role in the growth of the Internet and the boom of wireless communication channels.
Due in part to embedded systems, more and more consumer products and industrial equipment are becoming Internet-friendly. For example, DVD players are now dialling in to Internet databases for movie trivia, and GPS (Global Positioning System) mechanisms are often integrated into automobiles. It is all proof that the Internet and wireless technologies are not just for personal computers anymore.
Most embedded systems are small enough to sit on the end of your thumb and are usually hidden within much larger and more complex mobile computing or electronic devices, so they often go unnoticed. But embedded systems actually represent the vast majority of semiconductor sales. According to the World Semiconductor Trade Statistics blue book, there are an estimated 5 billion embedded microprocessors in use today — a whopping 94 percent share of the world market. (By comparison, unit sales of high-profile PC processors, such as the Intel Pentium and Motorola PowerPC, check in at only 6 percent market share.)
Embedded Systems Applications- Peaking into future
Embedded systems can be regarded today as some of the most lively research and industrial targets.
Embedded platforms were initially composed of separate discrete components. The processor was a separate component with just a memory bus interface, and all peripherals were attached to this bus. As integration levels increase, more and more logic is added to the processor die, creating families of application-specific service processors. The term system on chip (SOC) is often used to describe these highly integrated processors. These SOCs include much of the logic and interfaces that are required for a range of specific target applications. The silicon vendors that develop these SOC devices often create families of SOCs all using the same processor core, but with a wide range of integrated capabilities.
In this paper we discuss porting the Embedded Xinu operating system (OS) to the 8-bit Arduino Mega 1280 due to its ability to meet several constraints. A 64-bit Linux environment and associated tools are used to develop Xinu for the AVR. The steps to port Xinu can be split into several different steps. The final Xinu image for the Arduino is 35 KB in size without any compiler optimizations. We discovered interesting features of the platform such as the division of SRAM and are in a position to improve compatibility with existing Xinu files. However, there are many pieces of the Xinu for the AVR missing such as interrupt handlers, asynchronous serial driver, priority scheduling, and a better memory allocation function.
To meet and respond to its customers needs, IBM creates, develops and manufactures many of the worlds most advanced technologies, ranging from computer systems and software to networking systems, storage devices and microelectronics. Indeed, IBM has various product lines and services a few of which are: the Personal Computer that was first created in 1981, AS/400 business system, RS/6000 family of workstations and server systems, S/390 enterprise server, groundbreaking ThinkPad notebook computer; the award-winning IBM Netfinity and finally, PC Servers. It is an important supplier of hard disks, random access memories, and liquid crystal monitors.
Artificial intelligence, or AI, is a field of computer science that attempts to simulate characteristics of human intelligence or senses. These include learning, reasoning, and adapting. This field studies the designs of intelligent agents, or a system that acts intelligibly. The term artificial intelligence is confusing and misleading however. Artificial intelligence is still a form of intelligence, but perhaps “synthetic intelligence” is a better name because it is not natural intelligence. This is why the name “computational intelligence”, or CI, is sometimes preferred. Artificial intelligence is used in many objects that we use everyday: cars, microwaves, personal computers, and videogames.
Microprocessors and Angelic Self-possession: The microprocessors of today's computers are integrated circuits which contain the CPU on a single chip. The latest developments, with variable clock speeds now often exceeding 200 MHz, include Intell's Pentium chip, the IBM/Apple/Motorola PowerPC chip, as well as chips from Cyrix and AMD. The CPU chip is the heart of the computer; only memory and input-output devices have to be added. A small fan might be added on top of the fastest chips to cool them down, but in the chip itself there are no moving parts, no complex gaps between the movement being imparted and that which imparts the movement.
Embedded Computing: A VLIW Approach to design, Compilers and Tools- Joseph A. Fisher- Clifford Young - Paolo Faraboschi
A sensor node is also known as a mote (mostly in North America). This sensor node in a wireless sensor network is capable of performing some processing, gathering of sensory information and communicating with other connected nodes in the network. It must be clearly understood that a mote is a node but a node is not always a mote. Although wireless sensor nodes have existed for decades and used for applications as diverse as earthquake measurements to even warfare, the latest development of small sensor nodes dates back to the 1998 Smartdust project[1] and the NASA Sensor Webs Project[1]. One of the objectives of the Smart dust project was to create autonomous sensing and communication within a cubic millimeter of space. Though this project ended early on, it led to many more research projects. They include major research centers in Berkeley NEST [2] and CENS [2]. The researchers involved in these projects came up with the term mote to refer to a sensor node. The equivalent term in the NASA Sensor Webs Project[2] for a physical sensor node is pod, although the sensor node in a Sensor Web can be another Sensor Web itself. Physical sensor nodes have been able to increase their capability in conjunction with Moore's Law. The footprint of the chip contains more complex microcontrollers and these microcontrollers are also lower powered. Thus, for the same footprint of the node, more silicon capability can be now packed into it. Nowadays, motes or nodes focus on providing the longest wireless range upto dozens of km, with lowest energy consumption and the easiest development process for the user.
The perception layer includes a group of Internet-enabled devices that are able to perceive, detect objects, gather information, and exchange information with other devices through the Internet communication networks. Radio Frequency Identification Devices (RFID), cameras, sensors, Global Positioning Systems (GPS) are some examples of perception layer devices. Forwarding data from the perception layer to the application layer under the constraints of devices’ capabilities, network limitation and the applications’ constraints is the task of the network layer. IOT systems use a combination of short-range networks communication
System architecture and design of the system of an SoC is the most important parts to be considered when developing system-on-chip applications.
The Internet of Things (IoT) is the network of physical objects accessed through the Internet. The Network of interconnected objects harvests information from the environment (sensing) and interacts with the physical world (actuation/command/control), uses existing Internet standards to provide services for information transfer, analytics, applications and communications (Jiong et al. 2012).
With self-driving cars, facial-recognition software, mobile phones that listen to our very command and even robot-butlers, the future is now.
Artificial Intelligence (AI) is the intellect that unveiled by the devices. In computer science the supreme "intelligent machine” is a supple cogent factor that detects its surroundings and take actions that increase the chance of success in approaching the goals (Russell & Norvig 2003). In the colloquial manner, the term "artificial intelligence" is used when the machine imitate the intellectual feature that make able humans to
With the rapid development and constant evolution of systems technology to date, different development processes have been established. Each of these processes where designed with a specific design in mind but we will find out that many of them share common tasks and are geared towards achieving the same goal
According to the website Science Fiction and Fantasy World, artificial intelligence (A.I.) can be defined as "the simulation of human intelligence processes by machines, especially computer systems." (Moy, par. 2) This domain has predominantly been a field characterized by complex research in laboratory scale environments and just only recently has been becoming a part of the scene of technology in commercial applications. A.I. is about to become a real force in our technological evolution. There has been so much development and change in the last ten years that it is hard to believe how far we have come. But there is going to be another technical revolution going on, and it will be about A.I.
After two years working for Mobile Semiconductor, the only independent memory compiler vendor, I want to gain more knowledge in new edge-technology by returning to school. I have a passion of enhancing my professional career so I like to undertake some graduate courses in Embedded System at the University of Washington. This can keep me current with the latest technology in my chosen field and also support to develop a new area of expertise to further my professional career.