Marco Andreetto

A cooperative monitoring technique using visually servoed drones

This paper describes a distributed control and estimation technique aimed at optimizing the ability of a team of Unmanned Aerial Vehicles (UAVs) to detect and to locate a given target. The proposed algorithm relies on both visual servoing and a consensus scheme. Due to its flexibility, it can be potentially used in a variety of environmental, energy or structural monitoring applications. The general idea has been validated through simulations based on realistic parameter values.

Harnessing steering singularities in passive path following for robotic walkers

Assistive passive robotic walkers are naturally modelled as rear-driven bicycle with control on the front steering wheels. Standard path following algorithms used for unicycle-like robots can then be readily available, e.g. using backstepping techniques, to control the walker on desired paths. However, such an approach is usually singular in the very common situation of zero velocity, e.g. whenever the vehicle starts its motion or the user stops for any reason. The paper proposes a non-singular passive path following algorithm for an assistive robotic walker equipped with front steering wheels. The control law avoids the singularities, since it is velocity-independent, and allows the designer to specify saturation limits on the steering angles. The converging properties of the non-singular velocity-independent controller in the presence of saturation constraints are firstly formally proved and tested in simulations. Then extensive experiments performed on 14 testers are presented. These tests underline the promising performance of the proposed controller and the importance of singularities-avoidance in real-world scenarios to increase human comfort along the planned trajectory.

Path Following With Authority Sharing Between Humans and Passive Robotic Walkers Equipped With Low-Cost Actuators

A promising way to offset the reduced mobility of a large number of older adults is using robotic systems providing physical and cognitive support for the navigation in large and crowded public spaces. In this letter, we study a possible solution for guidance based on the action of electromechanical brakes mounted on the rear wheels. The solution does not require sophisticated sensing devices and aims to share the control authority between the user and the robotic platform: the system kicks in with quick corrections only in presence of deviations from the path or of dangerous situations. The level of the shared authority is controlled by means of a simple parameter. The control paradigm is based on an all-or-nothing policy relying on very rough actuators. In our experimental validation, all the users have been proficient in the use of the device after a short learning time.

Assistive robotic walker parameter identification for estimation of human thrust without force sensors

In this paper we propose a parameter identification procedure for the wheel-motor dynamic of a robotic walker, i.e. a commercial trolley for elderly people endowed with cognitive, sensing and guidance capabilities. The objective of the wheel-motor dynamic model is to generate a suitable time reference to be used in an estimation algorithm. The ultimate goal of the estimation algorithm is to retrieve the thrust, i.e. torque and force, that the older adult user of the robotic walker applies to the platform. These quantities are of paramount importance in order to adopt intelligent and comfortable walker guidance algorithms. The novelty of this approach is the avoidance of additional costly sensors that are usually used to solve such a problem. Experimental results are reported to show the effectiveness of the proposed approach.

Quasi time-optimal hybrid trajectory tracking of an n-dimensional saturated double integrator

Reference tracking is a challenging problem in the presence of saturation. This paper proposes a hybrid controller that ensures global convergence to zero of the error dynamics of a saturated multidimensional double integrator in the presence of a time-varying reference. Such a controller combines a local linear feedback, which ensures the tracking in the presence of small errors (local mode), and a global stabilizing quasi time-optimal controller, which handles large errors and the saturation (global mode). The proposed switching hybrid logic is applied also to the reference. In global mode, the reference trajectory is slowed down to promote the stabilization task. Once in local mode, the controller speeds up the reference to restore the nominal behavior. The proposed strategy has been extensively tested via simulations. The proposed solution has shown a graceful performance degradation in the presence of disturbances that do not allow the controller to remain in the local mode.