Summary:

This paper describes a way of optimization when designing robotic haptic interfaces. The 
performance measures were considered for both kinematic and dynamic criteria. The paper 
introduces the global isotropy index (GII) as a worst-case kinematic performance measure, 
and it is calculated from Jacobian matrix. A global dynamic index (GDI) is also 
introduced to quantify the global worst-case performance of a manipulator. GDI measures 
the largest effect of mass on the dynamic performance.

The problem was formulated with the objective of optimization to maximize GII while 
minimizing GDI at the same time. To solve the problem a branch-and-bound type algorithm 
was used for non-convex single objective optimization problems, and Pareto methods was 
chosen to characterize the trade-off between multiple design criteria.

A 2-dof five-bar linkage model was chosen to evaluate the suggested optimization 
framework. This test demonstrates the comparison between solutions from the suggested 
optimization framework and previously published methods.

Comments:

1. The paper is well organized and logical. It seems a very nice idea to construct a 
framework to optimize the design parameters in a robotic system.
2. However, there is not a novel idea or theory introduced in this paper. The paper 
chooses certain optimization techniques and applies them to specific haptic interface 
problems.
3. The paper provides too much information to explain optimization techniques and to 
introduce indices in the first two parts.
4. It would be better if the paper could demonstrate the suggested framework works for 
more general and complex cases.
5. Typo: "to chose" => "to choose" (The second line in the last paragraph in Section 6)