Paolo Fiorini

Motion Planning in Dynamic Environments Using Velocity Obstacles

This paper presents a method for robot motion planning in dynamic environments. It consists of selecting avoidance maneuvers to avoid static and moving obstacles in the velocity space, based on the cur rent positions and velocities of the robot and obstacles. It is a first- order method, since it does not integrate velocities to yield positions as functions of time. The avoidance maneuvers are generated by selecting robot ve locities outside of the velocity obstacles, which represent the set of robot velocities that would result in a collision with a given obstacle that moves at a given velocity, at some future time. To ensure that the avoidance maneuver is dynamically feasible, the set of avoidance velocities is intersected with the set of admissible velocities, defined by the robots acceleration constraints. Computing new avoidance maneuvers at regular time intervals accounts for general obstacle trajectories. The trajectory from start to goal is computed by searching a tree of feasible avoidance maneuvers, computed at discrete time intervals. An exhaustive search of the tree yields near-optimal trajectories that either minimize distance or motion time. A heuristic search of the tree is applicable to on-line planning. The method is demonstrated for point and disk robots among static and moving obstacles, and for an automated vehicle in an intelligent vehicle highway system scenario.

A Design and Control Environment for Internet-Based Telerobotics

This paper describes an environment for the design, simulation, and control of Internet-based force-reflecting telerobotic systems. We define these systems as using a segment of the computer network to connect the master to the slave. Computer networks introduce a time delay that is best described by a time-varying random process. Thus, known techniques for controlling time-delay telerobots are not directly applicable, and an environment for iterative designing and testing is necessary. The underlying software architecture sup ports tools for modeling the delay of the computer network, design ing a stable controller, simulating the performance of a telerobotic system, and testing the control algorithms using a force-reflecting input device. Furthermore, this setup provides data about including the Internet into more general telerobotic control architectures. To demonstrate the features of this environment, the complete proce dure for the design of a telerobotic controller is discussed. First, the delay parameters of an Internet segment are identified by prob ing the network. Then, these parameters are used in the design of a controller that includes a quasi-optimal estimator to compensate small data losses. Finally, simulations of the complete telerobotic system and emulations using a planar force-reflecting master and a virtual slave exemplify a typical design-and-test sequence.

Motion planning in dynamic environments using the relative velocity paradigm

A simple and efficient approach to the computation of avoidance maneuvers among moving obstacles is presented. The method is discussed for the case of a single maneuvering object avoiding several obstacles moving on known linear trajectories. The original dynamic problem is transformed into several static problems using the relative velocity between the maneuvering object and each obstacle. The static problems are converted into a single problem by means of a vector transformation, and the set of velocity vectors guaranteeing the avoidance of all the obstacles is computed. Within this set, the best maneuver for the particular approach can be selected. The geometric background of this approach is developed for both 2-D and 3-D cases, and the method is applied to an example of a 3-D avoidance maneuver.<<ETX>>

A Short History of Cleaning Robots

The definition of the desired functions and the design of an ultimate versatile personal robot is an ongoing debate. Meanwhile, however, precursors of this yet to evolve species are well on their way to become commercial products. Cleaning robots for public environments as well as for private households seem to be able to provide the breakthrough which the designers of non-industrial robot systems have long awaited.This survey describes a selection of 30 different cleaning robots, with the first developments reaching back more than 15 years. With a few exceptions we have focused on floor cleaning, in particular indoor floor cleaning. We describe a variety of scrubbing and vacuuming robots which were developed for this task. The described systems range from heavy, large, and expensive industrial cleaning vehicles to small-size, light-weight, low-cost household devices. Thesurvey does not include, however, systems for cleaning facades of buildings, or windows, or production tools.Although not all of the 30 cleaning robots abovementioned have yet reached the state of commercial products, their number alone certainly reflects the expectations regarding the economic value associated with the automation of cleaning tasks. In Europe only the estimates for the market for cleaning services range up to the order of US

Navigating a Robotic Wheelchair in a Railway Station during Rush Hour

100 billion per year. It is therefore not surprising that the cleaning industry and the manufacturers of cleaning devices arerather enthusiastic with respect to the automation of cleaning tasks using (semi-)autonomous mobile robot systems.

Time optimal trajectory planning in dynamic environments

This paper describes the hardware design, control, and navigation system of and some preliminary experiments with the robotic wheelchair Mobility Aid for elderly and disabled people (MAid). MAid’s general task is to transport people with severely impaired motion skills. The authors did not set out to reinvent and redevelop the set of standard skills of so-called intelligent wheelchairs, such as Follow Wall, FollowCorridor, PassDoorway, which are commonly described in the literature. These maneuvers require motion control skills that disabled people, in spite of their disabilities, are eager to learn and quite good at using. Instead, this work focused on generalizing the approach to fine motion control by considering those maneuvers identified as very burdensome due to their duration and required concentration. One of these functions is deliberative locomotion in rapidly changing, large-scale environments, such as shopping malls, entry halls of theaters, and concourses of airports or railway stations, where tens or hundreds of people and objects move around. MAid’s performance was tested in the central station of Ulm during rush hour and in the exhibition halls of the Hannover Messe ’98, the largest industrial fair in the world. Altogether, MAid has survived more than 36 h of testing in public, crowded environments with heavy passenger traffic.

Internet-based telerobotics: problems and approaches

This paper presents a method for motion planning in dynamic environments, subject to robot dynamics and actuator constraints. The time optimal trajectory is computed by first generating an initial guess using the concept of velocity obstacle. The initial guess, computed by a global search over a tree of avoidance maneuvers, is then optimized using a dynamic optimization. This method is applicable to repetitive tasks in known dynamic environments, as is demonstrated for a planar robot manipulator.

Health care robotics: a progress report

This paper describes the main issues facing the development of telerobotics systems connected to the Internet. The initial hurdle relates to the definition and implementation of an appropriate communication protocol for delivering real-time data. Two of the solutions proposed in the literature are briefly summarized. Then the paper examines the issues relevant to control systems in the presence of time delay, with particular reference to telerobotic systems with force feedback. The delay introduced by Internet connections have specific characteristics that are briefly summarized, together with the control theory approaches available to design stable algorithms. Finally, we present the results of experiments aimed at representing typical Internet connections in terms of parameters useful for the design of a control systems.

Issues on Internet-Based Teleoperation

This paper describes the approach followed in the design of a service robot for health care applications. Under the auspices of the NASA Technology Transfer Program, a partnership was established between JPL and RWI, a manufacturer of mobile robots, to design and evaluate a mobile robot for health care assistance to the elderly and the handicapped. The activities of the first phase of the project include the development of a multi-modal operator interface, and the design and fabrication of a manipulator arm for the mobile robot. This paper describes the architecture of the system, the features of the manipulator arm, and the operator interface.

Ground mobility systems for planetary exploration

Abstract This paper describes experiments of Internet modeling and of Internet-based teleoperation aimed at developing control laws to overcome the variable time-delay and data losses typical of Internet communication. From a control perspective, Internet is shown to be characterized by mean and jitter of the data packets delay, and by packet losses. Furthermore, these parameters depend on the number of nodes traversed by the packets and on the specific network traffic. The effects of packet delay jitter and losses on teleoperation performance are demonstrated using a 2-dof force feedback teleoperation master.