Image sensor initialization in a robotic vehicle

A method of image sensor initialization of an image sensor of a robotic vehicle, comprising:

• capturing an image with the image sensor of the robotic vehicle;
• extracting, by a processor, first features from the captured image;
• determining, by the processor, whether a number of extracted features exceeds a threshold number of features;
• maneuvering the robotic vehicle to another location in response to determining that the number of extracted features exceeds the threshold number of features;
• obtaining, by the processor, one or more of a translation of the robotic vehicle and a rotation of the robotic vehicle using a movement sensor of the robotic vehicle;
• determining, by the processor, whether an angular velocity of the robotic vehicle is less than an angular velocity threshold;
• matching, by the processor, one or more extracted first features with one or more second features extracted from a captured second image;
• estimating, by the processor, an image sensor pose based on matched features and one of the translation of the robotic vehicle and the rotation of the robotic vehicle;
• computing a first relative rotation matrix based on image information from the captured image and a second relative rotation matrix based on the one or more of the translation of the robotic vehicle and the rotation of the robotic vehicle obtained using the movement sensors of the robotic vehicle;
• and
• determining whether the first relative rotation matrix is approximately equal to the second relative rotation matrix.

A method of image sensor initialization of an image sensor of a robotic vehicle, comprising:

• capturing a first image with the image sensor of the robotic vehicle at a first location;
• extracting, by a processor, first features from the captured first image;
• extracting, by the processor, second features from a predefined image pyramid;
• matching, by the processor, one or more extracted first features with one or more extracted second features;
• estimating, by the processor, a first image sensor pose based on matched first features and second features and known dimensions of the target image;
• maneuvering the robotic vehicle to a second location and capturing a second image from the second location;
• extracting, by the processor, third features from the captured second image;
• maneuvering the robotic vehicle to a third location and capturing a third image from the third location;
• extracting, by the processor, fourth features from the captured third image;
• matching, by the processor, one or more extracted third features with one or more extracted fourth features;
• estimating, by the processor, a second image sensor pose based on the matched third and fourth features;
• determining, by the processor, that the robotic vehicle has returned to the first location based on a translation of the robotic vehicle and a rotation of the robotic vehicle;
• capturing a fourth image with the image sensor of the robotic vehicle at the first location in response to determining that the robotic vehicle has returned to the first location;
• extracting, by the processor, fifth features from the fourth image;
• matching, by the processor, one or more extracted fifth features and with one or more extracted second features;
• estimating, by the processor, a third image sensor pose based on the matched fifth features and second features;
• and
• determining the coordinate system for the robotic vehicle based on the first image sensor pose, the second image sensor pose, and the third image sensor pose.

A robotic vehicle, comprising:

• an image sensor;
• a processor coupled to the image sensor and configured with processor-executable instructions to:
• capture an image of a target image with the image sensor;
• extract first features from the captured image;
• extract second features from a predefined image pyramid of the target image;
• match one or more extracted first features with one or more extracted second features;
• estimate an image sensor pose based on matched features and known dimensions of the target image;
• determine whether the image sensor pose is within one or more thresholds;
• and
• in response to determining that the image sensor pose is not within the one or more thresholds:
• compute a preferred relative image sensor pose in response to determining that the image sensor pose is not within the one or more thresholds;
• determine a location associated with the preferred relative pose;
• and
• maneuver the robotic vehicle to the determined location from which to capture another image of the target image.

A robotic vehicle, comprising:

• an image sensor;
• a processor coupled to the image sensor and configured with processor-executable instructions to:
• capture a first image with the image sensor;
• extract first features from the captured image;
• determine whether a number of extracted features exceeds a threshold number of features;
• maneuver the robotic vehicle to another location in response to determining that the number of extracted features exceeds the threshold number of features;
• obtain one or more of a translation of the robotic vehicle and a rotation of the robotic vehicle using a movement sensor of the robotic vehicle;
• determine whether an angular velocity of the robotic vehicle is less than an angular velocity threshold;
• match one or more extracted first features with one or more second features extracted from a captured second image;
• estimate an image sensor pose based on matched features and one of the translation of the robotic vehicle and the rotation of the robotic vehicle;
• compute a first relative rotation matrix based on image information from the captured image and a second relative rotation matrix based on the one or more of the translation of the robotic vehicle and the rotation of the robotic vehicle obtained using the movement sensors of the robotic vehicle;
• and
• determine whether the first relative rotation matrix is approximately equal to the second relative rotation matrix.

A robotic vehicle, comprising:

• an image sensor;
• a processor coupled to the image sensor and configured with processor-executable instructions to:
• capture a first image with the image sensor at a first location;
• extract first features from the captured first image;
• extract second features from a predefined image pyramid;
• match one or more extracted first features with one or more extracted second features;
• estimate a first image sensor pose based on matched first features and second features and known dimensions of the target image;
• maneuver the robotic vehicle to a second location and capturing a second image from the second location;
• extract third features from the captured second image;
• maneuver the robotic vehicle to a third location and capturing a third image from the third location;
• extract fourth features from the captured third image;
• match one or more extracted third features with one or more extracted fourth features;
• estimate a second image sensor pose based on the matched third and fourth features;
• determine that the robotic vehicle has returned to the first location based on a translation of the robotic vehicle and a rotation of the robotic vehicle;
• capture a fourth image with the image sensor of the robotic vehicle at the first location in response to determining that the robotic vehicle has returned to the first location;
• extract fifth features from the fourth image;
• match one or more extracted fifth features and with one or more extracted second features;
• estimate a third image sensor pose based on the matched fifth features and second features;
• and
• determine the coordinate system for the robotic vehicle based on the first image sensor pose, the second image sensor pose, and the third image sensor pose.