Fig. 7: Comparison of the variability captured by each channel of the sensor (shown here in a top view) for the Baseline-4, Dense-Dataset, and Baseline-6 training sets. | npj Robotics

Fig. 7: Comparison of the variability captured by each channel of the sensor (shown here in a top view) for the Baseline-4, Dense-Dataset, and Baseline-6 training sets.

From: Let’s DENSE: a novel protocol for efficiently collecting dense and diverse data for tactile slip detection in robotic grasping

Fig. 7

The lighter the colour of a channel, the closer it is to the maximum standard deviation captured by the same channel across the three training sets. It can be seen that the channels for the Dense-Dataset are, generally, of a lighter colour than the remaining datasets, meaning that it contains a more uniform distribution of high standard deviations across the sensor surface.

Back to article page