Performance-based Trajectory Optimization for Path Following Control Using Bayesian Optimization

Abstract

Accurate positioning and fast traversal times determine the productivity in machining applications. This paper demonstrates a hierarchical contour control implementation for the increase of productivity in positioning systems. The highlevel controller pre-optimizes the input to a low level cascade controller, using a contouring predictive control approach. This control structure requires tuning of multiple parameters. We propose a sample-efficient joint tuning algorithm, where the performance metrics associated with the full geometry traversal are modelled as Gaussian processes and used to form the global cost and the constraints in a constrained Bayesian optimization algorithm. This approach enables the trade-off between fast traversal, high tracking accuracy, and suppression of vibrations in the system. The performance improvement is evaluated numerically when tuning different combinations of parameters. We demonstrate that jointly tuning the parameters of the contourand the low level controller achieves the best performance in terms of time, tracking accuracy, and minimization of the vibrations in the system.

Safe Zeroth-Order Convex Optimization Using Quadratic Local Approximations

“…For example, in any robot mission planning, the robot is placed initially at a safe point and needs to gradually explore the neighboring regions while ensuring feasibility of its trajectory. Similarly, in the optimization of manufacturing processes, often an initial set of (suboptimal) design parameters that satisfy the problem constraints are known [18].…”

Stress Flow Guided Non-Planar Print Trajectory Optimization for Additive Manufacturing of Anisotropic Polymers

“…Similarly, print trajectory planning for non-planar robotic deposition is studied in [20], ignoring, however, stress flow alignment and further manufacturability constraints for the printing; the same holds for [13]. Full end-to-end implementation for non-planar FFF optimization requires knowledge of the extrusion dynamics [21,22], machine kinematics [15,23], and efficient trajectory optimization formulations [24,23] that consider process constraints [15], material, printed geometry [25,26,27], and stress flow field [14]. Consequently, improving the non-planar printing process performance is not trivial and requires developing advanced methods to optimize print path trajectories for arbitrary geometries under a given stress flow field and material extrusion constraints.…”

Safe Risk-Averse Bayesian Optimization for Controller Tuning

“…BO-based tuning for enhanced performance has been demonstrated for cascade controllers of linear axis drives, where datadriven performance metrics have been used to intentionally increase the traversal time and the tracking accuracy while reducing vibrations [14]. In model predictive control (MPC), instead of adapting the controller for the worst-case scenarios, the prediction model can be selected to provide the best closed-loop performance by tuning the parameters in the MPC optimization objective for maximum performance [15]- [17].…”