Toward dynamical sensor management for reactive wall-following

Abstract

We propose a new paradigm for reactive wallfollowing by a planar robot taking the form of an actively steered sensor model that augments the robot's motion dynamics. We postulate a foveated sensor capable of delivering third-order infinitesimal (range, tangent, and curvature) data at a point along a wall (modeled as an unknown smooth plane curve) specified by the angle of the ray from the robot's body that first intersects it. We develop feedback policies for the coupled (point or unicycle) sensorimotor system that drive the sensor's foveal angle as a function of the instantaneous infinitesimal data, in accord with the trade-off between a desired standoff and progress-rate as the wall's curvature varies unpredictably in the manner of an unmodeled noise signal. We prove that in any neighborhood within which the thirdorder infinitesimal data accurately predicts the local "shape" of the wall, neither robot will ever hit it. We empirically demonstrate with comparative physical studies that the new active sensor management strategy yields superior average tracking performance and avoids catastrophic collisions or wall losses relative to the passive sensor variant. Abstract-We propose a new paradigm for reactive wallfollowing by a planar robot taking the form of an actively steered sensor model that augments the robot's motion dynamics. We postulate a foveated sensor capable of delivering third-order infinitesimal (range, tangent, and curvature) data at a point along a wall (modeled as an unknown smooth plane curve) specified by the angle of the ray from the robot's body that first intersects it. We develop feedback policies for the coupled (point or unicycle) sensorimotor system that drive the sensor's foveal angle as a function of the instantaneous infinitesimal data, in accord with the trade-off between a desired standoff and progress-rate as the wall's curvature varies unpredictably in the manner of an unmodeled noise signal. We prove that in any neighborhood within which the thirdorder infinitesimal data accurately predicts the local "shape" of the wall, neither robot will ever hit it. We empirically demonstrate with comparative physical studies that the new active sensor management strategy yields superior average tracking performance and avoids catastrophic collisions or wall losses relative to the passive sensor variant.

Research on Multi-sensor Cooperative Tracking Mission Planning of Aerospace Hypersonic Vehicles

“…At present the research about multi-sensors collaborative tracking mission planning focused on the design of collaboration algorithm, multi-sensors management and resources scheduling [5][6][7] . In reference [6] , the presented scheme that the conjunction of sensor management strategy and motion dynamics guarantees a superior tacking performance.…”

“…In reference [6] , the presented scheme that the conjunction of sensor management strategy and motion dynamics guarantees a superior tacking performance. And it has made significant progress in the field of robotic kinematics.…”

Active sensing for dynamic, non-holonomic, robust visual servoing

“…Its chief virtue here is to afford an immediate extension of the ISS framework [16] to goal sets which are compact manifolds rather than points [28] without the need for more intricate reasoning about the existence and nature of Control-Lyapunov Functions relative to non-goal components of the invariant set that would otherwise be required [29]. 3 However, in (11), that we will find it necessary to feed the task variables back into the θ dynamics, thereby imposing a coupling in the closed-loop. Let us define s := sin θ r , and pick…”

“…It is wellknown [8], [9], [10], [11] that simultaneous, coupled control of an actuated sensory subsystem in coordination with the robot's body dynamics can remediate the stabilizability of underactuated base platforms. Following suit by recourse to a panning camera, our unicycle-like base, the dynamical legged XRHex [12] re-implementation of RHex [13], raises Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USA.…”

Wall following and human detection for mobile robot surveillance in indoor environment

“…The antecedent in each fuzzy rule uses a type-2 fuzzy set to enhance the robustness [16]. Wall following is a method used by bugs and reactive mobile robot navigation to escape from local minima [1]- [7], [12], [18]. This research presents an autonomous wall following on indoor environment with the detection of human motion to support the mobile robot navigation task of surveillance using only local information of the environment gathered by sensors embedded on mobile robot.…”