We consider the case of robotic bin picking of reflective steel parts, using a structured light 3D camera as a depth imaging device. In this paper, we present a new method for bin picking, based on a dual-resolution convolutional neural network trained entirely in a simulated environment. The dual-r...

We consider the case of robotic bin picking of reflective steel parts, using a structured light 3D camera as a depth imaging device. In this paper, we present a new method for bin picking, based on a dual-resolution convolutional neural network trained entirely in a simulated environment. The dualre...

This work presents an NMPC (Nonlinear Model Predictive Control)-based algorithm for safe navigation of multi-agent dynamical systems in a variable environment with fixed or moving obstacles. The contribution lies in the construction of repulsive potential fields for the fixed and moving obstacles wh...

A control technique for counter-current and co-current guidance of underactuated marine vehicles is revisited and stronger stability properties are shown. In particular, the stronger property of uniform semiglobal exponential stability is shown for the complete multiple-equilibria closed loop system...

The corner cutting avoidance problem is an important but often overlooked part of motion planning strategies. Obstacle and collision avoidance constraints are usually imposed at the sampling time without regards to the intrasample behavior of the agent. Hence, it is possible for an agent to “cut the...

In this article we discuss some of the benefits of using an MPC as a mid-level controller between the path generator and the low-level joint controller of a robot system. The MPC handles rudimentary runtime constraints that are not considered during path generation. We compare two task space oriente...

Several design methods and principles have been proposed in the literature in order to guide the design of autonomous operations. Putting the required efforts into learning and using the methods is a daunting task, and experiences have shown that the use of methods meant to the help the design proce...

Several design methods and principles have been proposed in the literature in order to guide the design of autonomous operations. Putting the required efforts into learning and using the methods is a daunting task, and experiences have shown that the use of methods meant to the help the design proce...

In this article we present a reformulation of the invariance and immersion speed observer of Astolfi et al. as applied to mechanical systems with bounded inertia matrices. This is done to explore the possibility of its practical implementation e.g. for 6 degrees-of-freedom industrial robots. The ref...

In this article the model predictive path following controller and the model predictive trajectory tracking con-troller are compared for a robotic manipulator. We consider both the Runge-Kutta and collocation based discretization. We show how path-following can stop at obstructions in a way trajecto...