Hans P Moravec

7446766, Nov 4, 2008

Feb 8, 2006

11/350195

Hans Moravec - Pittsburgh PA, US

Seegrid Corporation - Lowell MA

G06T 17/00

345424, 345419, 345427, 345629, 352 57, 701211

A method of generating an evidence grid representing an environment comprises the steps of collecting a set of stereo images at multiple locations within the environment using stereo sensors, and processing the set of stereo images to determine occupancy and distance data associated with each point in space represented in the set of stereo images; applying a sensor model to each point in space to determine a probability of occupancy of each voxel representing each point in space; and generating an evidence grid of the environment by combining the probabilities for each voxel.

8427472, Apr 23, 2013

Nov 3, 2008

12/263983

Hans Moravec - Pittsburgh PA, US

Seegrid Corporation - Pittsburgh PA

G06T 17/00

345420, 345419, 345427, 345629, 382154, 435325, 701209, 701221

Mapping, localization, and navigation systems for use with a mobile unit are provided that use a sensor model with adjustable parameters. The system includes range sensors configured to collect range data and a data storage system having stored therein an evidence grid representing an environment and a sensor model comprising adjustable parameters representing inaccuracies of the range sensors. A grid engine adjusts the adjustable parameters based on received range data from the range sensors. A navigation module can direct the mobile unit though the environment using the evidence grid, received range data, and adjustable parameters. The grid engine can also locate the mobile unit within the environment by using the adjusted sensor model parameters to compare the received range data against the evidence grid.

8433442, Apr 30, 2013

Jan 28, 2009

12/361379

Scott Justin-Marl Friedman - Pittsburgh PA, US
Hans Peter Moravec - Pittsburgh PA, US

Seegrid Corporation - Pittsburgh PA

G06F 19/00
G05B 13/00

700245, 700275, 700259

Robots and methods implemented therein implement an active repurposing of temporal-spatial information. A robot can be configured to analyze the information to improve the effectiveness and efficiency of the primary service function that generated the information originally. A robot can be configured to use the information to create a three dimensional (3D) model of the facility, which can be used for a number of functions such as creating virtual tours of the environment, or porting the environment into video games. A robot can be configured to use the information to recognize and classify objects in the facility so that the ensuing catalog can be used to locate selected objects later, or to provide a global catalog of all items, such as is needed for insurance documentation of facility effects.

20090194137, Aug 6, 2009

Jan 28, 2009

12/361441

Scott Justin-Marl Friedman - Pittsburgh PA, US
Hans Peter Moravec - Pittsburgh PA, US

SEEGRID CORPORATION - Pittsburgh PA

A47L 9/28
A47L 5/00
G01C 21/00
G06F 19/00
B08B 5/04

134 18, 15319, 701200, 700245

In accordance with aspects of the present invention, a service robot, such as a robotic cleaner, can be configured to more effectively service an environment. The service robot can include one or more sensors that sense its location, the location of objects, or both, and can also include noise reduction elements. The service robot can determine that it is under a “furnishing” and implement a different servicing pattern.

20090198376, Aug 6, 2009

Feb 13, 2009

12/371281

Scott Justin-Marl Friedman - Pittsburgh PA, US
Hans Peter Moravec - Pittsburgh PA, US

SEEGRID CORPORATION - Pittsburgh PA

G05B 19/418

700248

A system is provided that includes at least one manager and one or more robots configured to communicate wirelessly. The manager can include certain functions that generate data, instructions, or both used by one or more robots. The manager can also facilitate communications among several robots, or robots could also be configured to communicate directly.

20090198380, Aug 6, 2009

Jan 28, 2009

12/361300

Scott Justin-Marl Friedman - Pittsburgh PA, US
Hans Peter Moravec - Pittsburgh PA, US
Mitchell Weiss - Carlisle MA, US

SEEGRID CORPORATION - Pittsburgh PA

G06F 7/00

700259, 700258, 901 1, 901 47

In accordance with aspects of the present invention, a service robot and methods for controlling such a robot are provided. In particular, the robot is configured to sense the presence of a person and to take a next action in response to sensing the presence of the person. As examples, the robot could leave the area, await commands from the person, or enter an idle or sleep state or mode until the person leaves.

20130201296, Aug 8, 2013

Mar 15, 2013

13/836619

Mitchell Weiss - Carlisle MA, US
Hans Moravec - Pittsburgh PA, US

H04N 13/02

348 48

A multi-camera head includes a plurality of stereo cameras mounted together and having their fields of view projecting in different directions, where the fields of view may or may not overlap. The cameras can be mounted to different sides of a frame and can be used in mobile contexts for mapping, navigation, or both, as examples. One or more of the cameras can be pitched at an angle, and a method may be provided for graphically viewing and analyzing field of view projections of the cameras considering the pitch angle.

55962602, Jan 21, 1997

May 13, 1994

8/242541

Hans P. Moravec - Pittsburgh PA
James R. Beninghaus - Cupertino CA
Michael K. Blackwell - Pittsburgh PA
Kevin J. Dowling - Pittsburgh PA
J. Rhoads Hollowell - Sunnyvale CA

Apple Computer, Inc. - Cupertino CA

H01M 1048

320 30

In an electronic device powered by a battery, a method of determining a charge of the battery. Initially, a discharge curve specifying the battery's voltage as a function of time is determined. This discharge curve is calibrated according to actual measurements. Next, a battery state model is established. The battery state model is comprised of a number of discrete charge states. The probability of the actual battery's charge for a particular charge state is specified by the battery state model for each of the charge states. The battery's voltage is periodically measured. Based on the measured voltage and the discharge curve, a voltage probability distribution is computed. The battery discharge model is updated by applying Bayes theorem to the old discharge model and the voltage probability distribution. The charge of the battery based on a mean value of the battery discharge model is then displayed.