The Robotic Vision Systems

Published Date: 02 Nov 2017

Executive summary

This report will primarily focus on the observation techniques in order to automatically gain knowledge of, recognition of environment and object relocation with the assistance of robotic vision. Huge component of the report focus on the advancement of methods to capture and process/develop the information that revolves around robots with the understanding that this information be capable of diversifying perception sensors as well as sensors concentrating on movements of robot such as stereo vision, ultrasonic sensors and color based video cameras. We as a group have also centered our focus on progression of vigorous techniques alternative to illumination, color, uncertainty of sensor, dynamic environment and reflectance towards surface. Future development and integrated application are also included with the report to portray and to be a sign of further opportunities that robotic vision can produce in various establishments and industries. Although in certain parts of the reports explains and moves to alternate methodologies, the information is integrated in similar perception formats, algorithms and image colorings. The diversification of in depth information will be clearly portrayed in order to clearly explain the sturdiness of the techniques. Limitations and problems of the robotic vision and surrounding the system is also described and identified in this report.

introduction

Since the remarkable significant success of the Apollo moon landings, advancement and experimentation on robotic engineering has become a huge priority of researchers in various field of expertise to expand the idea in order to minimize human role. The space scientist for instance has been reconsidering the need and importance of human functionality in space research and exploration due to the expansion of science and technology of robots and machines. Compared to a human labor driven industry, machines and robots has the ability to carry out assignments much faster, precise and accurately. The reliability of the robotic system is much more appreciated because tasks are executed according to the standard operating procedures (SOP) and hazard analysis critical control point (HACCP) of a particular assignment. Moreover it is a typical case in series production of applications. The robot systems are either customized or modified in lieu to the environment in relation of a permanent location or constructed in certain spatial relations to tally with a fixed environment.

Relevant effectors trajectory relevant to the task is essential to be planned precisely for the period of an offline phase and making sure that unanticipated events must be avoided or not happen during the online phase. In general, complicated awareness methods and techniques are deemed unnecessary in accordance to modified relations between the environment and robot system. As mentioned above, industrial robot system fundamentally operates predominantly in a well defined and constructed working environment with proper, complete controlled and documented operation conditions which mean the conventional robot system is a non adaptive system which is programmed to be utilized in standard conditions of the workplace and can be manipulated.

Some of the reason why the robot is manipulated and to be operated under conditional variation are object misplacements, the complexity of object or application, peculiarity of application and errors due to numerical quantization. The change in environment and condition causes the robot to not follow the appropriate modification trajectory. In the case of industrial robot for instance, there are various demanding applications where by the manipulator must maneuver a range of objects that are not specifically placed in exact location, defected objects must be rejected with the assistance of quality control and review or standard inspection procedure is essential to ensure excellence of object. The major part of this system will be the vision sensory system that are integrated with the robot or machine because it has the ability to replicate the vision of human sense which gives the opportunity for non-contact measurement of the surroundings. The goal in creating, exploring, advancing and continuous progress in robotic vision system is to ultimately increase flexibility, reduction of human labor, adaptability, minimization of production cost with current and future demanding applications.

Robotic vision system is developed and structured on the based on the optical recognition of objects that needs to be manipulated that are stationed in the workplace of the manipulator. The system is usually operated or controlled using an external computer that substitutes the controller of the industrial robotic system. The system is then operated with the desire to be on familiar terms with the objects which are positioned in workspace as well as to identify the accurate position and orientation of each meticulous object with the intention of leading the robot manipulator system. The robotic vision system has no constrains of usage in a particular industry or workspace because it can be used developed and structured to be utilized to control contrasting dynamic systems like for instance automotive vehicles, spacecrafts and even wielding machines.

The system is also primarily classified depending on the collective number of cameras and on its positions. Generally single camera vision is integrated due to the cost of installation and uncomplicated than multi camera vision system. The usual camera usage method used in the robotic systems is in-hand and out-hand. In – hand system is when the camera is fixed directly on the robot and out- hand or stand –alone is when the camera is used to observe the robot from certain distance. Both this integration system makes the computer vision problem much easier to focus and control. The visual servoing approaches are primarily classified on the reliance of design of the control proposal. There two main proposed schemes are essentially used with the dynamic system for visual servoing. One of the control scheme is known as direct visual servoing because the controller directly works out the input of the self motivated system whereby the vision is carried out at high speed (minimum 100Hz and at the rate of 10ms). The other control scheme of opposite to the the previous one known as indirect visual servoing due to the vision control calculates a reference control law that is transferred to the lower level controller of the self motivated system.

Present and Future

The robotic vision systems, algorithms, hardware and software trends focused above will primarily have continuous improvements and for sure intensify in the near future. Quicker hardware, much more intelligent tools and enhanced application software advancement and deployment conditions will facilitate a larger as well as deeper management of robotic vision industrialization. With the massive demand for expansion and research in the robotic field the, experts have seen positive directions in production pricing performance, sturdiness, and user friendly that is making once a dream to reality in distant future. Product manufacturers and research experts are now considering robotic vision as not only curiosity research but more towards a advanced tool for production automation.

Even though prospective users and enthusiast eagerly waiting for the future foreseeable new technology which incorporates quicker hardware and intellectual software, the current robotic vision technology is an indication of the future is now and it is an electrifying moment for robotic vision users and manufacturers. Robotic vision development gadgets will for sure provide a larger impact across various accessible and emerging target markets which can be categorized in the subsequent research fields : industrial, service in the domestic and professional sectors, surveillance security and robotic technology in space. While each of the research field have their own product specification and requirement demands, it is a vital necessity to identify similarities and connection of common challenges. This comes within reach of possibilities to classify, gather, assess and review the main technology requirements to accomplish necessities which will directly set aside assessment of the timely feasibility of future manufacturing.

Contemporary robotic vision manufacturing is reasonably inflexible. On average, robots/machines are programmed and left to work for long periods of time on a particular operation/functions. In countries that the government and private sectors are facing relatively high manual labor cost and deficiency of skilled workforce, the population will continuously rely on robotic and robotic vision system in the industrial and professional service surroundings. With the integration of robotic vision system, further hazardous and unhygienic occupation will be carried out by robots which in long term will consequence in expansion of more bio knowledge/humane based employment descriptions. This has to be accepted by the majority as the only way to continue mass production, reconstruction, maintenance and preservation in a competitive and demanding world.

In the near future, robots with multiple cameras and image capturing will eventually be assembled and integrated with much complex job specification for instance combination of utilizing several arms, adaptive capabilities to execute multiple jobs, predominantly with the assistance of human intervention but later in parallel/automatic. Robots will be able to execute task that are much larger scale for example bridges and submarines and even in smaller scale such as micro and nano scale. Robots in due course assist humans in various state of affairs with close relation will demand compatibility to attain a mush safer and reliable operation in workspace, public, household and even space. Robot vision systems will also be used to the well being of human and/or equipments such as assisting medically enable patients in hospitals or at the privacy of patients home.

This sort of systems in security sector and also be a huge benefit for space programs which assist in eradicating large sums for expensive gears and risky space walks. Robotic vision can be identified in current transportation system in developed nations, for example automated train transits for passengers at the airports and in the future significantly reducing mobility requirements to hospitals and public domains subsequently increase the efficiency of road, sea and air transportation. Robotic vision development in the surveillance and security industry will integrate with situational attentiveness and decision making providing services in public areas, home and border protection both working on ground, water or in the air. In the long run, with the assistance of vision, robots will be able to respond to impulsive, sudden and unpredicted complex situations which could also be relatively dangerous and life threatening.

Multifaceted border protection missions will significantly increase and obligate deployment and collaboration of numerous robotic systems. Robotic vision will be a huge advantage for operations in areas that are both difficult to get to or hazardous to humans which can underground mining, war zone, demolished building structure and space exploration. In present robots are usually tele-operated and/or its autonomy is constrained to a restricted figure of steps. In the near future, autonomy level has to be elevated not limited in domains of communication but also in space research and exploration during time crucial processes. Such action can be accomplished with the collaboration of multiple robotic vision systems.

Researchers are already in the process of developing robotic vision system in educational aids, fitness training, gaming companions and motion simulators whereby these robots will have constant communication/interaction with a physical and mental level. They will assist in educating, playing games with children and social companionship for senior citizen and disabled person. Robotic vision in the future will be able to convert and interpret human emotions and physical expression in order to provide the necessary assistance. Industrial robotic sectors in Europe, North America, Russia and Japan has become a strong competitive industry, with the expansion of their current success into other areas will assist in eradicating the gap involving industry and academic through widespread technological transfer and network. The main issue and challenge with the target market is to manufacture hardware and software with adequate functionality to create attraction and fascination to generate interest of public over a time period at the appropriate price for the large demanding market.

Conclusion

The utmost challenging prospect in robotic vision system is the infusion and integration of various technologies from a wide range of element and fundamental domains into one coherent system. Robotic vision system will often frequently seek their way into daily usage devices. The expansion and advancement of computer engineering skills, techniques, methods and tools are equally important for such development to take place now and the near future. As mention above in the report, producing, manufacturing and crafting robotic vision systems will directly maximize productivity, reliability and quality of objects which in return eradicates stressful and dangerous working environment. Robots in service industry will contribute and assist in our day to day standards of living as well as providing border control protections. The awareness and knowledge of robotic vision development among people has increased the demand with non technical problems that could possibly turn out to be a barrier to mass market which indirectly forces to develop awareness at the beginning level in conjunction with the technological research expansion. Standard Operating Procedures (SOP) and policy makers must engage with researchers, industrialist and common users/operator to establish frameworks towards safe and responsible operation. Decision making robotic vision systems must be embedded and Integrated with behavioral ethics and safety procedures to avoid hazardous practice at workspace.

Robotic vision researchers and experts and trained users are fundamentally required for purposes such as designing, integrating, research and development of both hardware and software. Some of the limitations that will cause the development to hold back would be moderate skilled employees, lack of resourses in various field of expertise, physics and computer science. One of the best opportunities for further development would be to focus on the technologies that are needed across different field in the future and create additional benefit by reusing and researching technologies unconditionally. The leading limitation to the development of robotic vision system for both researchers and enthusiast is funding or sponsor. Large sums capital need to be invested to set up the appropriate surrounding to conduct research and development for any field to examine and expand.

Fundamentally, support a team of researchers with equipments, machineries and superior computerized technology will cost funder and sponsor large amount of capital which directly constrains the idea of expansion in certain field of work. Shipping and travelling cost for the research team are also significantly high since large number of associates will be required to run such assignments. This does not limit to the maintenance of the robotic vision hardware and software without sponsor from private sponsorship participant attraction.

Last but not least, accurately structured and designed robotic vision system will definitely encourage the progression of robotic future. Even with all the innovations that are being developed at current time, to be competitive and booming further, current system needs copious alternations with the purpose of immense impact for future generation.