Paulo Gurgel Pinheiro

Anticipative Shared Control for Robotic Wheelchairs Used by People with Disabilities

Patients with neurological disorders are often required to cope with off-putting and longer rehabilitation process and some of them become wheelchair users with severe restriction on the arms and hands skill. Many studies have proposed solutions for autonomous or semi-autonomous powered wheelchairs focusing on navigation problems. In this project, we are interested in bringing up a planning shared control model for wheelchair to be used by people with disabilities or inexpert users to move from a room to another into an indoor place. The main objective is to reduce the effort of the users on controlling the wheelchair on these dedicated paths. This leads us to improve an algorithm of shared control to use other data sources such as agendas, travelled path, pose of the wheelchair, and time of day to infer patterns of conduction, in order to anticipate the conductors intentions, for example, his/her intended path or final destination and to communicate with the user providing smart suggestions during the process. This anticipative shared control is of prime importance when tiring human-machine interfaces such as BCI, eye-tracking or EMG are being employed.

Planning for multi-robot localization

This paper will present a cooperative multi-robot localization model with planning support. Models of communication and transmission of pose estimates are constantly explored, however how the robots act on the environment is generally defined by random actions (from the localization tasks point of view). Random actions generate observations that can be useless for improving the estimate. This work describes a proposal for multi-robot localization with planning of actions. The objective is to describe a model where policies define the best action to performed by robots. The proposed model, called Model of Planned Localization - MPL, uses POMDPs to model the problems of location and specific algorithms to generate policies. We compared the MPL to a model that does not make use of planning actions. The results showed that MPL is able to estimate the positions of robots with lower number of steps, being more efficient than model compared.

Arrangement map for task planning and localization for an autonomous robot in a large-scale environment

This paper presents a planning approach for solving the global localization problem using an arrangement of rooms to compress the original map. The approach is based on architectural design features of the building such as walls and doors to help the robot on finding the best route to go. Lighter POMDP plans are generated only for representative rooms of the environment, decreasing size of the set of possible states. The plans are created offline only once and used indefinitely regardless of missions combining them online. The plan only requires as input, the environment map and the robot actions and possible observations. We demonstrate the single level approach and the map decomposition with experiments on both V-REP Simulator and the Pioneer 3DX robot. This approach allows the robot to perform both the localization and tasking in a large-scale environment.

A novel platform supporting multiple control strategies for assistive robots

This work presents a platform for the development of a functional prototype for assistive robotic vehicles supporting various control strategies in the context of a smart environment. The implemented framework allows an operator with a disability to interact with a smart environment by means of handfree devices (small movements of the face or limbs through Electromyograph (EMG), or Electroencephalograph (EEG), among others). The present work also details the integration and testing of four control strategies (manual control, shared control, point to go, and fully autonomous), giving the user the opportunity to choose among them based on the structure of the environment, personal preference, or capability. An intelligent assistive agent was integrated into the framework which helps the operator navigating the user interface and interacting with the environment. The controls performances for a common scenario are compared to validate the platform and compare the implemented navigation algorithms, and experimental results are presented and discussed.

Laser based driving assistance for smart robotic wheelchairs

This paper is presenting the ongoing work toward a novel driving assistance system of a robotic wheelchair, for people paralyzed from down the neck. The users head posture is tracked, to accordingly project a colored spot on the ground ahead, with a pan-tilt mounted laser. The laser dot on the ground represents a potential close range destination the operator wants to reach autonomously. The wheelchair is equipped with a low cost depth-camera (Kinect sensor) that models a traversability map in order to define if the designated destination is reachable or not by the chair. If reachable, the red laser dot turns green, and the operator can validate the wheelchair destination via an Electromyogram (EMG) device, detecting a specific group of muscles contraction. This validating action triggers the calculation of a path toward the laser pointed target, based on the traversability map. The wheelchair is then controlled to follow this path autonomously. In the future, the stream of 3D point cloud acquired during the process will be used to map and self localize the wheelchair in the environment, to be able to correct the estimate of the pose derived from the wheels encoders.