Autonomous mobile robots typically require a preconceived and very detailed navigational model (map) of their intended operating environment, but most law enforcement and urban warfare response scenarios preclude the availability of such a priori information. ROBART III is an advanced demonstration platform for non-lethal robotic response measures, incorporating a supervised autonomous navigation system specifically configured to support minimally attended operation in previously unexplored interior structures. A 'human-centered mapping' strategy has been developed to ensure valid first-time interpretation of navigational landmarks as the robot builds its world model. The accuracy of the robot's real-time position estimation (and hence the model itself) is significantly enhanced by an innovative algorithm which exploits a heuristic that assumes the majority of man-made structures are characterized by parallel and orthogonal walls. Intruder detection and assessment capabilities are supported by intelligent fusion of data collected by a multitude of various sensors. Initial detection is by a 360-degree array of eight passive-infrared motion detectors responding to the thermal energy gradient created by a moving human target, with partial validation from a Doppler microwave motion detector. The robot's head-mounted sensors (and non-lethal weapon system) are then panned to the center of any perceived disturbance for further assessment. Automatic tracking of any confirmed movement is accomplished through conventional image processing using a black-and-white video surveillance camera equipped with a near-infrared illuminator for low-light conditions.
Supervised Autonomous Security Response Robot
1999
18 pages
Report
No indication
English
Personnel Detection , Pattern Recognition & Image Processing , Bionics & Artificial Intelligence , Military Sciences , Emergency Services & Planning , Robots , Thermal properties , Moving targets , Position finding , Self operation , Autonomous navigation , Intrusion detection , Automatic tracking , Motion detectors
Reverse Optical Flow for Self-Supervised Adaptive Autonomous Robot Navigation
British Library Online Contents | 2007
|Design of an Autonomous Exterior Security Robot
NTIS | 1994
|Design of an autonomous exterior security robot
AIAA | 1994
|Design of an autonomous exterior security robot
NTRS | 1994
|