Wyniki wyszukiwania - Biblioteka Nauki

1

Interactive mapping using range sensor data under localization uncertainty

100%

Vieira P. , Ventura R.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2013

|

tom Vol. 7, No. 1

47--53

EN

Several methods have been proposed in the literature to address the problem of automatic mapping by a robot using range scan data, under localization uncertainty. Most scan matching methods rely on the minimization of the matching error among individual range scans. However, uncertainty in sensor data often leads to erroneous matching, hard to cope with in a purely automatic approach. This paper proposes a semi-automatic approach, denoted interactive mapping, involving a human operator in the process of detecting and correcting erroneous matches. Instead of allowing the operator complete freedom in correcting the matching in a frame by frame basis, the proposed method facilitates the adjustment along the directions with more ambiguity, while constraining the others. Experimental results using LIDAR data are presented to validate empirically the approach, together with a preliminary user study to evaluate the benefits of the approach.

2

Absolute localization for low capability robots in structured environments using barcode landmarks

100%

Dias D. , Ventura R.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2013

|

tom Vol. 7, No. 1

28--34

EN

Research on multi-robot systems often demands the use of a large population of small, cheap, and low capability mobile robots. Many non-trivial behaviors demand these robots to be localized in real-time. This paper addresses the problem of absolute localization of such low capability robots using onboard sensors and local computation. The approach is based on the use of a pair of scan lines perceived by an onboard B&W camera to detect and decode artificial visual landmarks deployed along the environment. Each landmark consists on a dual-layer barcode which can encode its pose with respect to a global coordinate frame. Thus, the robot is not required to store a map of the landmark locations onboard. The method is based on an Extended Kalman Filter (EKF) fusing odometry readings with absolute pose estimates obtained from the camera. Experimental results using an e-puck robot with 8 KB of RAM and a 16 MIPs processor are presented, comparing the location estimates with both ground truth and odometry.

3

Assisted teleoperation of quadcopters using obstacle avoidance

100%

Mendes J. , Ventura R.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2013

|

tom Vol. 7, No. 1

54--58

EN

Teleoperation of unmanned aerial vehicles often demands extensive training. Yet, even well trained pilots are prone to mistakes, resulting frequently in collisions of the vehicle with obstacles. This paper presents a method to assist the tele-operation of a quadrotor using an obstacle avoidance approach. In particular, rough map of the nearby environment is constructed using sonar sensors. This map is constructed using FastSLAM to allow tracking of the vehicle position with respect to the map. The map is then used to override operator commands that may lead to a collision. An unknown and GPS denied environment is considered. Experimental results using the USARsim simulator are presented.

4

Simultaneous localization and mapping for tracked wheel robots combining monocular and stereo vision

100%

Jesus F. , Ventura R.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2013

|

tom Vol. 7, No. 1

21--27

EN

This paper addresses an online 6D SLAM method for a tracked wheel robot in an unknown and unstructured environment. While the robot pose is represented by its position and orientation over a 3D space, the environment is mapped with natural landmarks in the same space, autonomously collected using visual data from feature detectors. The observation model employs opportunistically features detected from either monocular and stereo vision. These features are represented using an inverse depth parametrization. The motion model uses odometry readings from motor encoders and orientation changes measured with an IMU. A dimensional-bounded EKF (DBEKF) is introduced here, that keeps the dimension of the state bounded. A new landmark classifier using a Temporal Difference Learning methodology is used to identify undesired landmarks from the state. By forcing an upper bound to the number of landmarks in the EKF state, the computational complexity is reduced to up to a constant while not compromising its integrity. All experimental work was done using real data from RAPOSA-NG, a tracked wheel robot developed for Search and Rescue missions.