New framework jointly optimizes robot inspection route and arm configuration

A new paper from Minh Nhat Vu and colleagues introduces a unified framework for multi-viewpoint robotic inspection that jointly optimizes the visiting order and per-viewpoint inverse-kinematics (IK) configuration. Unlike traditional modular pipelines that fix a single IK solution per viewpoint, build an all-pairs travel-time map, and then route independently, this method encodes the three-dimensional self-motion manifold of each viewpoint in closed form, ensuring pose constraints are satisfied by construction. A derivative-free optimizer (CMA-ES) minimizes a cheap penalized objective over both order and configuration, after which direct-collocation trajectory optimization is applied only to the edges of the selected route—reducing trajectory solves from quadratic to linear in the number of viewpoints.

In simulations and real-robot experiments on a KUKA LBR iiwa with a 2-DoF linear stage, the approach demonstrated feasibility, smooth execution, and reduced end-to-end inspection time compared to modular and naive distance-based baselines. The key innovation is that the rest-to-rest travel time between viewpoints is approximated using a closed-form admissible double-integrator surrogate, and the tour is encoded via random keys. This removes the decoupling that prevents modular pipelines from achieving global time optimality, making the framework particularly valuable for high-precision industrial inspection tasks where minimizing cycle time is critical.