Tactile Ergodic Coverage on Curved Surfaces

TL;DR

<3-5 sentence high-level summary> This work addresses the challenge of tactile coverage on curved surfaces under complex contact dynamics by proposing a closed-loop ergodic control framework that operates directly on point clouds. The approach combines diffusion-based ergodic planning with a conformal geometric algebra–driven task-space impedance controller to simultaneously track a surface-normal line and apply a controlled contact force. It leverages a spectral diffusion formulation for real-time computation and validates the method via simulated and real-world experiments on kitchenware, demonstrating robust tactile coverage and promising avenues for tactile data collection. The results indicate the method's practical potential for autonomous tactile exploration and dataset acquisition on non-planar geometries.

Abstract

In this article, we present a feedback control method for tactile coverage tasks, such as cleaning or surface inspection. These tasks are challenging to plan due to complex continuous physical interactions. In these tasks, the coverage target and progress can be easily measured using a camera and encoded in a point cloud. We propose an ergodic coverage method that operates directly on point clouds, guiding the robot to spend more time on regions requiring more coverage. For robot control and contact behavior, we use geometric algebra to formulate a task-space impedance controller that tracks a line while simultaneously exerting a desired force along that line. We evaluate the performance of our method in kinematic simulations and demonstrate its applicability in real-world experiments on kitchenware. Our source codes, experimental data, and videos are available as open access at https://sites.google.com/view/tactile-ergodic-control/

Figures (10)

• Figure 1: Overview of our feedback control method for tactile coverage. Left: We measure the surface and the red target using the camera and encode them in a point cloud. Bottom-right: We diffuse the target and use its gradient field to guide the coverage. Then, we close the loop by measuring the actual coverage with the camera and use it as the next target. Top-right: We measure the tactile interaction forces using the force sensor and the tool orientation using the joint positions. We solve the geometric task-space impedance control problem using a line target and a force target along the line.
• Figure 2: Blue-red points show the value of the potential field $u_\tau$ on the pointcloud $\mathcal{P}$ and the yellow point is the projected agent position $P'_a$. We also project the agent's neighbors $P_i$ to the tangent plane $E_{a,\tau}$, shown in green. Next, we use the height function $h_i=u_{i,\tau}$ which uses the values of the potential field to lift the projected points in the normal direction of the tangent plane. We show the lifted points with large blue-red points. We fit a polynomial to this lifted surface and compute its analytical gradients at the neighbor locations $\nabla u_{i,\tau}$, as shown with arrows in the detail view.
• Figure 3: Information flow between the three components. The pipeline is composed of an outer loop responsible for controlling the coverage progress with the feedback from the camera, whereas the inner loop compensates for the mismatch due to the robot dynamics.
• Figure 4: Computational complexity of the preprocessing step for different $n_{\mathcal{P}}$ and $n_{M}$. Legend shows $n_{M}$ values. The time axis is logarithmic and the legend shows $n_{M}$ values.
• Figure 5: Computational complexity of integrating the diffusion equation at runtime for different $n_{\mathcal{P}}$ and $n_{M}$. The time axis is logarithmic and the legend shows $n_{M}$ values.