Report on Robotic grinding in manufacturing

As stated by Ding et al. (1), the multi-axis tools of CNC Machine as well as the virtue operation majorly grind the complex components currently. The mainstream approach for the manufacturing of the parts has been become by the multi-axis CNC grinding because of the harsh operating environment experienced and the time as well as the labour consuming (2). Although, in the depth of the application that is limited in arrears to the following reasons such as the high cost of the exactness machine tools (3), which are most expensive and has a high millions UA dollars specifically for the large-scale multi-axis CNC machine tool, the complex configuration without integrated function of the measurement of the machining as well as the characterized and the manufacturing mode that is fixed by the parallel machining capability as well as the unavailable flexible (4).

            The new ideas are offered by the approach which based on industrial robots for the production of the very complicated components. The robots are looking attractive because of the competitive price as well as the large extendable workspace by comparing with the multi-axis tools of the CNC machines that makes them a cost effective solution to match up the complicated components and especially for those parts which are comparatively have large dimensions (5). The running parameters within the real time based on the information of multi sensor feedback as well as the process knowledge model can be optimized by operations of the robotic machining, the force sensing as well as the machine vision, specifically armed with the powerful sensing functions. The break by using the boundaries of the equipment for the manufacturing traditionally that emphases on just the axis position of the movement as well the controlling on the speed, thus leading for the active control of the equipment on the process. Therefore, it is significant for mentioning that the major barrier for vast usage of the robots within the exactness of the machining is their repeatability as well as the low accuracy for the tools of the CNC Machines as well as a very brief analysis of the tools of the CNC machine as well as those robots which are used for the industry for the machining (6). However, it has proposed a large number of the impactful solutions for the reduction of the positioning the errors as well as the manipulator stiffness within the industrial machining fields (7). The robotics grinding is progressively doing replaced the tools of the multi-axis of CNC machine, as well as the becoming the unconventional part means for the manufacturing. The results of the research of the robotic grinding of the complicated components are progressively augmented during the previous one or two periods, as well as the published journals articles majorly focus on the feasibility study of the robotic machining and the designing or modelling as well as the comparison of the dynamics of the machining (8). It mainly conducts the investigations from the aspects of the robot the grinding path planning, robot calibration as well as the measurement, the control of the force, the robot posture or the position optimization as well as the control of the removal. Furthermore, it is also found, the current ones are majorly expounded for the perspectives of machining robots as well as the properties of the machining which is based on the mobile robot machining, the categories of the operation as well as the robotic chatter machining (9).

Challenges faced by robotic grinding of Robotic grinding in manufacturing

In the manufacturing sector, the significant parameters which tend to serve as the main hurdles to the robotic grinding are as given:

Acquiescence concerns: Other than the precision and accuracy related challenges, which take place at the system level, the components of the robotic grinding systems are also of immense importance. These components are a part of the geometrical accuracy of the system. It is mandatory to consider the compliance control due to its impact on the contact state along with the pressure influence for the purpose of having the control over machining allowance (5). The consistent power control procedures can be condensed into four classes: impedance control, cross breed power/position control, versatile control as well as canny control. There is huge advancement to the practices of robotic systems for the grinding purposes. Still, the force control is less in practical engineering applications.

This is predominantly attributable to the troubles in precise demonstration of mechanical elements and usage of power control calculations, which are essentially obliged by the transparency of business modern robots. In the interim, the current power/position control in mechanical crushing expects to lessen the surface harshness of parts and gives less consideration to the exactness of shape and position. Indeed, for some perplexing segments, for example, turbine sharp edges, they have unpleasantness necessities, yet additionally providing with the prerequisites for forming the exactness.

For robotic grinding systems in the manufacturing industry, the major challenge regarding the high-precision force/position control is recorded. It is about uncovering the power/position coupling in two symmetrical subspaces. The aim behind it is to provide the guarantee for the power controlled ordinary bearing in case of contact-force. Also the big question is to deal with guarantee for providing the direction which is to follow precision in force-controlled unrelated course?

Precision concerns: For the current research work, the precision concerns of the robotic grinding systems in the manufacturing industry are considered to be the biggest challenge. For the robotic grinding systems the efficiency as well as the accuracy is desirable. This phenomenon is all about measuring the area related to the performance of the work. After doing this, the pin-points need to be identified as per already available theoretical models. By doing this, the details related to the corresponding position are better & extracted quickly (7).

The main purpose for this evaluation is the cross-check for both the designed models as well as the measured points. It is done with the help of performing the calculations for the parameters related to the rigid transformation. The major challenges that prevail to the way of performing such calculations may include as given: the accuracy as well as the precision cannot be determined regarding the absolute positioning. The complicated components cannot be scanned properly based on the existing robotic systems for the manufacturing industry.  The significant elements for affecting the favourable rigid transformation may include the local data missing, Gaussion noise, as well as the uneven point compactness. It tends to provide the failure to the traditional algorithm for matching purposes.

The components over the surface tend to provide with the incorrect values for the robotic grinding in manufacturing. It may include the profile errors as well as the allowance distribution problems. It can be said the designed models as well as the measured points need to be cross-checked with the intense accuracy & the precision. It is basically perceived to be the key parameter for making the robotic grinding systems to perform accurately along with their position points.

Accommodating concerns: In case the unpredictability of the machining task enhances, it tends to offers the constraints for the machining capacity of a grinding robot for the manufacturing sector. The transactional relevancy helps the robotic grinding structures in the manufacturing industry to work as a whole i.e., by making use of different components. For the said purpose, the components are perceived to be the varying operators which are to assist the proper functionality of the grinding robots. It better helps to indicate the points of interest in effectiveness, adaptability and adaptability, particularly for huge scale complex structures (2).

It can be said that for the manufacturing industry, the major components of the grinding robotics are discernment, multi-sensor information combination, multi-robot design, and resolution of the conflict. For the case of multi-robot agreeable control, the conventional strategies partition the entire framework into various slave-master frameworks and perform remote facilitated control on every subsystem. By doing this it is tried to bring the improvements to the dependability and synchronization of numerous frameworks through the cooperative control of Internet model.

It is hard to acquire the worldwide data of the work piece, be that as it may be, for the synchronous machining of huge scale complex structures. Also, the requirement connection between different robots is mind boggling and hard to tackle precisely. It tends to introduce the interest on the constant detecting and data move of numerous robots. What's more, it is better suggested to advance robot movement control techniques. These are appropriate for shared machining, and build up the controller just as community oriented control programming to acknowledge task designation and obstruction evasion (10).

            Below is given the flow chart for model processing for grinding robots in the manufacturing industry.

Figure 1: Flow-chart for model processing for grinding robots in the manufacturing industry

Typical applications of robotic grinding for the complex components

A. Use of turbine blades for Robotic belt grinding in manufacturing

a. Background along with the existing problems

Currently, the ideals of manual grinding as well as multi-pivot CNC belt pounding help to process the cutting edges which is related to the most part wrapped up by Contrasted and the sharp edge profile. If we talk about the thickness of the main and trailing edges of the cutting edge then we find it is the thinner one as given: (to be R0.1mm level). Both the bending as well as the machining way change enormously. It represents a test to the exactness granulating of sharp edges. While for the multi-pivot CNC grinding, the cutting edge clipping situating mistake is equal to the disfigurement blunder of the sharp edge. It is related to the request for size of the said system. So, it is required that CNC grinding must be founded on the exact estimation of the edge clasping state (11).

Be that as it may, the machining method of the current multi-hub CNC pounding hardware is fixed (for example uninvolved granulating, no adaptability and parallel preparing capacity). Combined with the perplexing arrangement, it is hard to shape the mix of "measurement-machining". The creators' exploration group collaborated with Wuxi Turbine Blade (WTB) Limited Company to research the automated belt crushing of different sorts of steam turbine sharp edges and flight edges, and made extraordinary enhancements in both machining effectiveness and surface quality.

For instance, the hour of mechanical belt granulating one meter long steam turbine cutting edge was abbreviated to a little ways from the 45 minutes by manual pounding activity, the precision of the edge profile was expanded from ±0.15mm to ±0.1mm, and the surface harshness was improved to Ra0.4 from Ra0.8 correspondingly. In any case, it was found from the long haul nearby tests that mechanical belt granulating of complex sharp edges is essentially looked with the accompanying issues:

·         Unevenly disseminated pounding expulsion remittance. Most importantly, the cutting edge expulsion stipend isn't equally dispersed before pounding, and this is basically identified with the past procedure. For instance, in the wake of manufacturing the air motor blower sharp edges, just the cutting edge roots, canals and driving and trailing edges should be ground and cleaned, and the machining remittance dispersion at the main and trailing edges is very uneven from 0.042 to 0.224 mm. Besides, the sharp edge bracing blunder additionally influences the machining remittance circulation of the edge driving and trailing edges. Due to the unevenly conveyed machining stipend and the thin machining data transmission, the "unfilled running marvel" may happen in the edge area of the cutting edge, and the uneven pounding power irritates the babble, in this manner bringing about the granulating marks which is likewise significant to the firmness of robot-clip sharp edge framework (8).

·         Difficulty in controlling the crushing contact power. From one viewpoint, the guideline of consistent weight crushing is normally used in rough belt pounding, yet this strategy is clearly hard to adjust to the edge evacuation of uneven machining edges. In the interim, the pounding procedure joined by the wear and strip off of abrasives, makes the thickness of the rough belt change. Step by step instructions to guarantee the exact evacuation of the cutting edge from the procedure puts appeal on the contact power control. Then again, since the main and trailing edges of the sharp edge are thin, the crushing is inclined to torsional and bowing distortion (9).

To guarantee the exactness of the cutting edge profile, a little contact power underneath 10 N is carefully required. Be that as it may, on account of adaptable contact with clear flexible shape, how to keep exorbitant disfigurement blunder from the granulating hardware additionally puts a pressing requirement for contact power control.

Strategies for successful robotic grinding

Some strategies for the successful robotic grinding of the complicated components are discussed here, which are explained in the further paragraphs. It can conduct the following effective strategies as well as the potential solutions for the construction of the combined measurement, manipulating the machining function to overcome the challenges confronted by the complicated components of the robotic grinding for the robotic grinding system.

1). Process Parameter optimization of Robotic grinding in manufacturing

The process parameter optimization is used for the improvement of the integrity surface of the workpiece robotically machined. Cycle-time and belt oscillation frequency the researchers did analyze the effects of two significant process variables in belt finishing, on the forming surface as well as specificities finish of steel that is hard by an approach of energetic tribo. A systematical investigation on floor integrity of nickel-based first-rate alloy in robotic belt grinding used to be carried out from the aspects of morphological structure, floor roughness, residual stresses, and structural area size. The consequences of abrasive grain size, contact force, linear belt speed and feed fee on the surface roughness in abrasive belt grinding of aviation blades, and the most efficient manner parameters were decided by means of examining the response surface. The outcomes indicate that the continuous partial dynamic recrystallization coupled with mixed outcomes of plastic deformation and thermal therapy contributes to the excessive qualitative floor integrity, and each grinding pressure and belt pace are discovered to appreciably influence the floor integrity.

The machining as well as the flexibility contact along with the wide strip that are two prominent compensations or the pros for the system of the robotic belt grinding as well as it may be widely used for the improvement of the quality of the surface as well as the effectiveness of the robot but the pieces of the work with the statue of the curvature (12). The running parameters within the real-time based on the information of multi-sensor feedback, as well as the process knowledge model,  can be optimized by operations of the robotic machining, the force sensing as well as the machine vision, specifically armed with the powerful sensing functions (13). Therefore, it is significant for mentioning that the major barrier for vast usage of the robots within the exactness of the machining is their repeatability as well as the low accuracy for the tools of the CNC Machines as well as a very brief analysis of the tools of the CNC machine as well as those robots which are used for the industry for the machining.

The research on the grinding planning path for the robotic machines may be found by the study of the kinematics of the contact. The universal demands of the technologies of the grinding of the belt must be satisfied by the processes of the grinding, as well as the most significant thing is to form the contact wheel imitate the features of the geometrical on the contact local area (14). The results of the research of the robotic grinding of the complicated components are progressively augmented during the previous one or two periods, as well as the published journals articles majorly focus on the feasibility study of the robotic machining and the designing or modelling as well as the comparison of the dynamics of the machining (15).

The length of the curve among the cutting locations of the neighboring develops the longer for ensuring the efficiency of the processing. Furthermore, to make sure or investigate the accuracy of machining (13), the curve length becomes shorter for the local areas along with the larger curvature. For the intersection along with the target surface of the section for grinding, a series of planes are also developed as well as the corresponding curves of the social corresponding are obtained. The length of the curve among the cutting points of neighboring for every curve is optimized by the addition of the cutter location at the local area along with those curvatures which are looking big the size. The paths of the grinding generation method consist of the length spacing optimization is shaped (14). The validity of the design, structure as well as the process is completely approved by the offline simulation as well as the surface quality during the experiments of the grinding with the process is enhanced. The theoretical support for the smoothness, as well as the path of the robotic grinding of the surface’s accuracy is provided by the method of the planning of the path. (14).

2). Machining error Compensation of Robotic grinding in manufacturing

As the authors, Chen & Dong (3) described that the robotic machining was reported in early studies within the 1990s. Eventually, the potential of robot applications in the machining has realized that there are a lot of researches on the machining of robots available in the shape of series worldwide. The researchers Chen & Dong (3) in this paper will take a look at the modern development of the machining of robot. Such development works may be roughly classified within the proposed researches on the system development of the robot machining, the compensation, stiffness of the modeling or dynamic, path planning of robot machining, as well as the chatter analysis or vibration analysis together with the path tracking (16). Such kind of researches on the robot machining will clearly improve the efficiency as well as the accuracy of robot and for development of such kind of robot machining systems, provision of the useful references for the tasks one time thought to be capable by CNC machines (17).

As described by the authors Chen & Dong (3), for the advancement of the technology of the robotic machines to the competitiveness of the systems as well as the next level, which is more complicated and practical could be developed. So that’s why the researchers (3) propose that the next studies on the machining of robots must have to focus on the comparison of the robotic machines, the difficulty in planning of the path based on the map, the optimization of the robotic link of arm, scheduling as well as the planning for a line of machining robots.

Since the stiffness of the robotic grinding setup was at least one order of magnitude lower than established grinding machines, the complete gadget stiffness such as a tool, workpiece, and robot stiffness in robotic grinding operation is regarded to be a determination that motives the deviation of machining accuracy. Focus on the huge impact of tool deflection on the machined surface in robotic grinding, the work derived a novel grinding force model and real-time tool deflection 21 compensation algorithm to effectively classify the grinding operation regimes and predict the grinding forces (15).

Critical issues and Methodologies

As Zhang et al. (12) stated as this research is providing brief information about critical issues as well as some methodologies for the improvement in the machining performance of robotics along with those robots that are flexible in the industry. The stiffness of the robots is importantly lower as compared with the CNC machine while in the outcomes it has low productivity as well as unacceptable quality. Furthermore, the problem is treated with the methodology of the novel, which includes the real-time deformation compensation for the quality, the stiffness of modeling as well as the controlled material removal rate (12) to make the system efficient. The experimental outcome shows that better surface accuracy, as well as higher productivity, may be attained, indicating a promising as well as the use of the practice of industrial robots to applications, which is not possible at the current time (4).

Zhang et al. (12) described that to minimize the machining blunders generated via the no uniform material removal in belt grinding, proposed a second-order osculation primarily based method for grinding marine propeller blades, and the outcomes confirmed that the machining error could be controlled below the given tolerance. Particularly in belt grinding of blades, the abrasive belts with distinct materials under the motion of bendy contact wheel would undergo giant elastic deformation, and this will reason deformation error of the blade area processing (16), and finally have an effect on the blade profile accuracy and the surface quality. Actually, the effect of elastic deformation at tool-workpiece interface in abrasive belt grinding is extra sizeable and was once appreciably investigated by using setting up an average contact pressure mannequin to evaluate the friction coefficient and abrasive wear rate (12).

Wu et al. (18) stated that the model of the systematic geometry had been presented to the standardization of a modern designed the blade of 5-axis turbine robotic machine for grinding. To get higher level efficiency as well as the great accuracy for grinding of the turbine blades, through the removal of the process of hand grinding, this machine is designed (18). Furthermore, the design of the machine is totally different from the tools of the competitive machine, the topology for that is RPPPR in which R is rotary while P is prismatic. To identify by investigating, maintaining, managing as well as improving the accuracy of the calibration, recommended for the acceptance testing as well as the performance assessment, an error quantification is the only way.

The error modeling technique in the symmetric geometry is applied as well as the dependant position, as well as the position of the independent errors, is recognized but the machine consideration machine as the stiff structure of the robots by removing the errors of setup of the cutting tool as well as the piece of work. Moreover, thirty-nine of them are identified for having the effective errors, as well as these, are suggested to get the total effect among the piece of work as well as the cutting tool within the volume of the workspace. It uses the homogenous transformation matrices as well as the techniques of the rigid body kinematics (18).

Material removal rate Modelling of Robotic grinding in manufacturing

As stated by Bernd & Xiang (12), the key indicators for uring the profile with the accuracy as well as the accuracy in the dimensions of surfaces of the machine within the robotic grinding are the material removal rate. The premise for the high performance of robotic grinding, such as the conventional operation for wheel grinding is to clarify the mechanism of the removal of the material meanwhile it is largely influenced by the properties of the material, the information of the geometry as well as the parameters process.

As described by Bernd & Xiang (12), since the cloth removal is not definitely a geometric computation in robotic belt grinding, later work more focuses on its theoretic modeling, which is seen as a contact problem between an elastic body and a rigid body. The researchers considered the fabric removals as the pressure distribution in the contact area and put ahead an SVR-based nonlinear force calculation model primarily based on the regular global grinding model. Then the real-time simulation of the robot-controlled belt grinding strategies will become possible (12).

Particularly for the abrasive belt grinding, the pioneered work 14 experimentally investigated the outcomes of contact pressure between tool and workpiece, belt pace and feed fee on the fabric removal fee in the course of robotic belt grinding of turbine-vane overhaul, and found that the normalized fabric elimination amount used to be elevated with the make bigger of the contact load and belt speed, and with the limit of the feed rate (12).

 High-Quality Belt Grinding System of Robotic grinding in manufacturing

As stated by Shuihua & et al. (13), robotic belt grinding is very useful technique to make the pieces of work along with the complex free-form geometries. Therefore, the control strategies as well as more sophisticated modeling because of the minimum stiffness in the system are called for. A novel model for the approximation of material removal is presented in this literature within the grinding process of robotic belt. Two process parameters, such as the contact force as well as the velocity of the robot among the contact wheel and the piece of work are analyzed within the process model. Furthermore, for the estimation of the distribution of the pressure, (13) a superposition technique is also introduced in the contact area. The existing method highly decreases the time for the computation as compared to Finite Element Analysis (FEA) method as well as also gives the equation explicitly to analyze the real-time system.

Moreover, on material removal, a shape-dependent model is also developed for the estimation of the removal of material or material removal. To donate the ability of the material removal system in particular areas, the model presents the local coefficients. Simply, the proposed methodology may adapt necessarily for the prices of work along with the complex geometry, while the experimental outcomes confirmed the accuracy as well as the effectiveness of the model (13).

4). Contact Force Control of Robotic grinding in manufacturing

Figure 2: Association between components of the grinding robots in the manufacturing industry

An inexorably significant utilization of modern grinding robots is in the zone of manufacturing, for example, crushing as well as the manufacturing processes. For the said processes the contact force control of the robots is related with the procedure of the metal expulsion. It is perceived to be a significant thought. As per the available literature, there exist the variations regarding the contact force control strategies with the perspective of the grinding through robotics in the manufacturing industry. It can be either in the form of contact force control through the adjustments of the robots having simple as well as the open loop positioning or “through the arm” robotic control. The former will provide with the passive-end effector and the later will provide with the active end effector. It is due to the technological advancements to the robotic systems. In the said domain, a lot of work has been performed in order to introduce the advanced methods for the robotic force controllers.

For current type of contact force controls, the firmness, latency, as well as mechanical framework’s damping can better be transformed. It helps to manage the robot's reaction for various info collaboration powers with respect to the manufacturing processes in the manufacturing industry. In most of these examinations, in any case, controls for the dynamic power are better executed. It does not include any type of criticism regarding the metal evacuation procedures. The use of the plate granulating is not only popular for the manufacturing field but also for the car foundries, car businesses as well as the delivery yards etc. The main reason behind using this process is the assistance for the evacuation of deformities, scales along with the welding dots from the manufacturing machinery (1).

For the purpose of adjusting the position of the robot, which is designed for providing with the robotic grinding in the manufacturing, a power controller better helps to serve this objective. The main motive for making use of the power controllers include as given:

·         The rate for the metal evacuation is predetermined. By making use of the contact force control, the rate for the metal evacuation can be maintained.

·         The geometric blunders, which take place due to the contact force control, are reduced to the maximum extent.

·         In case of the cutting processes, this control provides with the guarantee regarding the forestall work-piece bum.

In the manufacturing industry, the use of the robotic grinding is suggested. The contact force controllers are helpful for managing the granulating power within the sight of a stage aggravation in the organizational stature. Notwithstanding, the reason given for the said purpose includes as given: the presentation of the contact force control, in the working area, is exceptionally delicate for the robotic arm area. Versatile control was seen as important to make up for the varieties for the parameters of the framework. The purported "long-extend prescient controllers" are fit for the elite systems as well as the applications. For example, the control of adaptable automated frameworks can be considered in this regard. Specifically, it is an extremely appealing control procedure for machining applications. For example, in the crushing it is necessary that power direction can be resolved during the undertaken arrangements.

It is also mandatory to carry out the procedures with the consideration of the assurance of the quality. This is better done by following up the processing cycle as given below: It can be seen that the cycle takes into consideration both the process adoption as well as the quality assurance. The process adoption is all about the automation of the process adaption along with the generation of the partial programs for the processing of the materials and the substances.

The high-quality of the processes and the outcomes is better ensured by selecting the technical parameters of the robots for the grinding purposes in the manufacturing industry. It is not enough to carry out the procedures rather the quality assurance is also as significant as the process itself. It is done by identification of the defects through well-defined evaluation processes. It can be done by adopting both either the manual classification of the defects or the automated identification of the same (6).

References of Robotic grinding in manufacturing

1. Adaptive Force Control for Robotic Disk Grinding. Elbestawi, M. A., Yuen, K. M., Srivastava, A. K., & Dai, H. 1, 1991, CIRP Annals, Vol. 40, pp. 391–394.

2. Tri-Co Robot: a Chinese robotic research initiative for enhanced robot interaction capabilities. Ding, H., et al. 6, 2018, National Science Review, Vol. 5, pp. 799-801.

3. Robot machining: recent development and future research issues. Dong, Yonghua Chen & Fenghua. 9-12, June 2013, The International Journal of Advanced Manufacturing Technology, Vol. 66, pp. 1489–1497.

4. Machining with flexible manipulator: toward improving robotic machining performance. Zhang, Hui, et al. 2005, Proceedings, 2005 IEEE/ASME International Conference on Advanced Intelligent Mechatronics.

5. Modeling of the working accuracy for robotic belt grinding system for turbine blades. Qi, J., et al. 6, 2017, Advances in Mechanical Engineering, Vol. 9, pp. 1-12.

6. A Robot System for High Quality Belt Grinding and Polishing Processes. Kuhlenkötter, Bernd & Zhang, Xiang. 2005, Science Direct, pp. 755-770. 10.5772/4680.

7. Calibration and accuracy analysis of robotic belt grinding system using the ruby probe and criteria sphere. Xu, X., et al. 2018, Robotics and Computer-Integrated Manufacturing, Vol. 51, pp. 189-201.

8. 3-D Shape Matching of a Blade Surface in Robotic Grinding Applications. Li, W.-l., et al. 5, 2016, IEEE/ASME Transactions on Mechatronics, Vol. 21, pp. 2294-2306.

9. Effect of the belt grinding on the surface texture: Modeling of the contact and abrasive wear. Jourani, A., et al. 7-12, 2005, Wear, Vol. 259, pp. 1137-1143.

10. RTRobMultiAxisControl: A Framework for Real-Time Multiaxis and Multirobot Control. Fischer, H., et al. 3, 2019, IEEE Transactions on Automation Science and Engineering, Vol. 16, pp. 1205-1217.

11. Equivalent self-adaptive belt grinding for the real-R edge of an aero-engine precision-forged blade. Xiao, G. and Y. Huang. 9-12, 2015, The International Journal of Advanced Manufacturing Technology, Vol. 83.

12. A Robot System for High Quality Belt Grinding and Polishing Processes. Zhang, Bernd Kuhlenkötter & Xiang. 07 01, 2005. 3-86611-038-3.

13. A MATERIAL REMOVAL MODEL FOR ROBOTIC BELT GRINDING PROCESS, Machining Science and Technology. Wu Shuihua, Kazerounian Kazem, Gan Zhongxue & Sun Yunquan. 18, 2014, Vol. 1, pp. 15-30.

14. A Path Planning Method for Robotic Belt Surface Grinding. WANG, Wei and YUN, Chao. 4, 2011, Vol. 24, pp. 520-526.

15. Automated polishing process with a human-like dexterous robot. . Y. Takeuchi, D. Ge & N. Asakawa. May 1993, IEEE International conference on robotics and automation, pp. 950–956.

16. Haptic-aided robot path planning based on virtual tele-operation. XJ He, YH Chen. 4-5, J Robot and Comput Integr Manuf, Vol. 25, pp. 792–803.

17. Robot-polishing of curved surface with magneto-pressed tool and magnetic force sensor. Nakagawa, M. Kunieda & T. 22–24, Apr 1985, Twenty-Fifth International Machine Tool Design and Research Conference; Birmingham, Vol. UK, pp. 193–200.

18. Systematic Geometric Error Modeling for Workspace Volumetric Calibration of a 5-axis Turbine Blade Grinding Machine. Wuyi, Abdul Wahid Khan & Chen. 5, 2010, Vol. 23, pp. 604-615.