N. Le Fort-Piat

CAD Model-based Tracking and 3D Visual-based Control for MEMS Microassembly

This paper investigates sequential robotic microassembly for the construction of 3D micro-electro-mechanical systems (MEMSs) structures using a 3D visual servoing approach. The previous solutions proposed in the literature for these kinds of problems are based on 2D visual control because of the lack of precise and robust 3D measures from the work scene. In this paper, the relevance of the real-time 3D visual tracking method and the 3D vision-based control law proposed is demonstrated. The 3D poses of the MEMSs are supplied in real-time by a computer-aided design model-based tracking algorithm. This algorithm is sufficiently accurate and robust to enable a precise regulation toward zero of the 3D error using the proposed pose-based visual servoing approach. Experiments on a microrobotic setup have been carried out to achieve assemblies of two or more 400 μm × 400 μm × 100 μm silicon micro-objects by their respective 97 μm × 97 μm × 100 μm notches with an assembly clearance from 1 μm to 5 μm. The different microassembly processes are performed with a mean error of 0.3 μm in position and 0.35×10 −2 rad in orientation.

Distributed control architecture for smart surfaces

This paper presents a distributed control architecture to perform part recognition and closed-loop control of a distributed manipulation device. This architecture is based on decentralized cells able to communicate with their four neighbors thanks to peer-to-peer links. Various original algorithms are proposed to reconstruct, recognize and convey the object levitating on a new contactless distributed manipulation device. Experimental results show that each algorithm does a good job for itself and that all the algorithms together succeed in sorting and conveying the objects to their final destination. In the future, this architecture may be used to control MEMS-arrayed manipulation surfaces in order to develop Smart Surfaces, for conveying, fine positioning and sorting of very small parts for micro-systems assembly lines.

Planning robust displacement missions by means of robot-tasks and local maps

Abstract This paper introduces a mobile robot mission planning method based on the robot-task concept. Missions are performed by a sequence of closed-loop mechanisms and are thus more robust with respect to the unavoidable differences between reality and the model us the planner. A displacement mission is defined as the combination of a path and of a description of the observations expected along the way, enabling thus to control/check its correct execution. These observations are obtained by the use of local maps defined as coherent sets of landmarks or beacons detectable by the on-board sensing devices. The considered vehicle is a car like one endowed with a rotative telemeter. The environment is 2D and is structured by polygonal obstacles and landmarks. Robot-task generation is carried out in three phases. The first phase consist of defining a path as a set of configurations. Each of these configurations is attached to the list of map primitives providing the robot with the best localization information. This part is a compromise between length, safety conditions and complexity of vehicle control. The second phase applies a data analysis method to the pain, in order to dynamically group path configurations ssociated to the same list of primitives called local maps. The third phase precisely deals with robot-task definition. It first consists in defining a smooth path (Bezier curve) from each group of configurations. The robot-task sequence correspond to the ordered list of these pathsassociated with their respective local maps.

Hybrid behavioral control architecture for the cooperation of minimalist mobile robots

This paper presents a hybrid control architecture based on subsumption and schemas motors principles in order to achieve complex and cooperative tasks. The control architecture implemented is constituted by a set of independent and elementary behaviors organized in layers of skills. Specific low-level behaviors, called altruistic behaviors and inspired by societies of insects (attractive or repulsive signals), are used to improve the efficiency of the control. Therefore, competitive and cooperative mechanisms are used in a unique hybrid architecture of control to perform a complex box-pushing task by a set of mini-robots. The analysis of an elevated number of simulations allows us to have statistical results (time to complete the task was chosen as performance criteria) which show the existence of an optimal number of robots to achieve the box-pushing task and underline the importance of the use of altruistic behaviors to enhance the cooperative task.

Automatic micromanipulation using multiscale visual servoing

The paper deals with the manipulation of silicon microcomponents in order to assembly automatically. The size of the components vary from 600 mum times 400 mum times 100 mum to 300 mum times 300 mum times 100 mum with a notch of 100 mum thickness on every side. The microassembly process is split up into elementary tasks (aligning component, positioning component, centering component, opening gripper, ...) where every one is achieved by visual servoing. The control laws are of the type exponential or polynomial decrease of error according to the task. The performing of the latter has required the implementation of an effective tracking algorithm in combination with a depth-from-focus technique in order to maintain the target in focus and to recover the distance between the gripper and the component. The process includes the adaptation of the video microscope magnification to the required resolution (coarse to fine servoings). A multiple scale modelling and calibration of the photon video microscope is performed. The picking and placing of above components were achieved : the errors of positioning are respectively 1.4 mum in x and y and 0.5 degree in orientation.

Improving Reinforcement Learning Speed for Robot Control

Reinforcement learning (R-L) is an intuitive way of programming well-suited for use on autonomous robots because it does not need to specify how the task has to be achieved. However, RL remains difficult to implement in realistic domains because of its slowness in convergence. In this paper, we develop a theoretical study of the influence of some RL parameters over the learning speed. We also provide experimental justifications for choosing the reward function and initial Q-values in order to improve RL speed within the context of a goal-directed robot task

Safe actions and observations planning for mobile robots

Deals with the definition of safe tasks for displacement missions of a non-holonomic mobile robot in an indoor environment. The goal of this study is to plan actions (path following) as well as observations (local maps), leading the robot to configurations where pertinent features can be sensed, thus allowing the robots best localization relative to its environment. Firstly, the local map concept is introduced as a set of the best landmarks used for planning and executing robust motion movements. Secondly, a path planning method dealing with uncertainties is proposed, where both uncertainties in localization and in control of a non-holonomic mobile robot are managed. The safeness of the proposed method is due to the mixing between the planning phase and the navigation phase.

A safe and robust path following planner for wheeled robots

This paper addresses the problem of planning a path following, robust with respect to small localization errors, allowing a mobile robot to reach safely its goal. A path planning method dealing with uncertainties is proposed where both uncertainty in localization and in control of a nonholonomic mobile robot are managed. The safeness of this method is due to the combination of the planning phase and the navigation phase by using the same localization and control processes. The localization process realized the multisensor fusion of data provided by a ring of telemetric sensors in combination with a dead-reckoning system. The robustness is due to a stabilizing controller on which the system remains stable when estimated values are used for feedback instead of exact ones. By determining a compact attractive domain bounding the regulation error, we dynamically compute a security margin during the path planning. This security margin ensures the safeness of the planned path.

Evolutionary Parameters Optimization for an Hybrid Control Architecture of Multicriteria Tasks

This paper presents a methodology for the determination of optimal parameters for a hybrid behavioral control architecture, which uses both, subsumption and schema motor principles, in order to achieve complex and multicriteria cooperative tasks. The proposed architecture of control is constituted by a set of independent and elementary behaviors organized in layers of skills. Specific low-level behaviors, called altruistic behaviors inspired by societies of insects (attractive or repulsive signals), are used to improve the efficiency of the control to perform cooperative box-pushing task. The parameters optimization are obtained using genetic algorithms. Appropriate genetic operators are used to manipulate real chromosomes which are constrained to have its genes with specific values, this constraint is dictated by the proposed control architecture. The validation of the results of optimization is established using a large number of simulations

Driving Activity: How to Improve Knowledge of the Environment

This paper concerns the problem of driving assistance and, in particular, how to improve the perception of the surrounding environment to make this assistance really helpful. The main aims of a Driving Assistance System are to improve the security of the driver, passengers, and other road users. Driving is a complex activity, where the interactions between the driver, the vehicle, and the environment are continuous and numerous. The vehicle moves in a dynamic environment, so the Driving Assistance System, for its diagnosis, needs a map that represents as well as possible the actual situation of this environment. This paper presents a multi-sensor fusion module embedded in a real vehicle. The problem considered here is the dynamic reconstruction of the environment of the vehicle, based on measurements of a set of sensors.