Dear Akarsh,

could you please be a little more specific?  There are so many different types of kinematic chains and control strategies.  I do not believe that there is any such thing as THE best and newest design.  But if you have a specific application and perhaps a manipulator topology in mind, there are probably a lot of people out there that can comment on the design steps you might want to take.

Regards,

Bjorn

I agree with Bjoem anad Igor.

When you speak about the cost, many parameters should be known. How many axis, the wieght, hight, number of motors, the weight of objects etc. Contolling is the last thing to talk about in cost. For examples sevo motors that carry 4KG/CM costs 10$ however, 20KG/CM costs 50$.

I'm really sorry for late reply. I have been looking into the designs myself. One thing I have noticed is that mostly all arms use sensors like accelero-meter, flex sensor, etc. Now the design of using servo motors in arm movement is too main stream. I have attached the image here. So I want something which is currently in research using a different implementation. Maybe it uses different sensors or a new mechanical design of arm itself. The arm should contain 5 DOF (Degree of Freedom). Also speaking of the cost, the budget is around $300. So, implementing which has already been implemented has no point so I'm asking for a newer mechanical design.

Also speaking of the weight lifting, it should be capable of lifting only a handful of light objects like a simple ball, a can, a pencil, etc.

I want a design which is more or less similar to a real arm and imitates the movement of how you move your arm.

I hope I have made it more comprehensive.

The designing activity, for a robotic arm, starting from beginning or for an implemented model, is depending on CAD software, of computeers and the number of designe engineers, of complexity of the robotic arm.

Dear Akarsh,

thanks for the extra details you have mentioned above. 

If you seek to "imitate" a human arm, you will likely be faced with choosing 7 DOF (all rotary), which does not include any of the DOFs in the hand. 

Concerning sensors, you might need to consider joint angle sensors, which are of course quite the usual way of doing things - yet I don't see a viable alternative if you seek to control positions. 

There may be an alternative to servo motors, but the reason these are used successfully today is the result of a development over many decades.  Long ago, before the mid-1980s, one used DC motors.  Given faults in the drive system, it was possible for these to run off leading to unexpected and dangerous motion.  With the PWM (pulse width modulation) driving of servo motors, it is really unlikely for a single failure in the control or drive system to give rise to unexpected motion.  Such failures will usually lead to a standstill of the motor.  In others words, today's servo drive systems are much safer than at least one alternative. 

Many ways of driving articulated robot arms have been studied, including cable drives for example that mimic the human system of tendons, or pneumatic artificial muscles that are modeled after human muscles.  Such alternatives are very interesting from a research point of view, but I do not expected these to be relevant in industrial practice anytime soon. 

Good luck with your work!

Cheers

Björn

Firstly,  I would like to apologize for a late reply. And I appreciate your help a lot. I will look into the videos and also look into the models mentioned above.

Thank You Once Again.