Abstract: A device and method for multimodal localization and mapping for a mobile automation apparatus is provided. Features are extracted from images acquired by a ceiling-facing image device of the mobile automation apparatus in an environment, and stored in association with estimated positions of the mobile automation apparatus in the environment as determined from one or more sensors, as well as in association with features extracted from depth data acquired from a depth-sensing device, for example as map data. The map data is later used by the mobile automation apparatus to navigate the environment based, at least in part, on further images acquired by the ceiling-facing image device.

Abstract: A device and method for multimodal localization and mapping for a mobile automation apparatus is provided. Features are extracted from images acquired by a ceiling-facing image device of the mobile automation apparatus in an environment, and stored in association with estimated positions of the mobile automation apparatus in the environment as determined from one or more proprioceptive sensors, as well as in association with features extracted from depth data acquired from a depth-sensing device, for example as map data. The map data is later used by the mobile automation apparatus to navigate the environment based, at least in part, on further images acquired by the ceiling-facing image device.

Abstract: The present invention provides a guidance, navigation, and control method and system for an underground mining vehicle that allow said vehicle to be taught a route by a human operator and then have it automatically drive the route with no human intervention. The method works in three steps: teaching, route profiling, and playback. In the teaching step the vehicle is manually driven by a operator along a route which can consist of an arbitrary sequence of maneuvers including tramming forwards, switching directions, tramming backwards, turning, or pausing movement. During this phase raw data from vehicle-mounted sensors including odometric sensors and rangefinders are logged to a file throughout teaching for later processing. During the (offline) route profiling step, the raw data in the log file are processed into a route profile, and a profile of desired speed as a function of distance along the path.

Abstract: The present invention provides a guidance, navigation, and control method and system for an underground mining vehicle that allow said vehicle to be taught a route by a human operator and then have it automatically drive the route with no human intervention. The method works in three steps: teaching, route profiling, and playback. In the teaching step the vehicle is manually driven by a operator (or using tele-operation whereby the operator views a screen displaying live views from vehicle-mounted cameras and using remote controls) along a route which can consist of an arbitrary sequence of maneuvers including tramming forwards, switching directions, tramming backwards, turning, or pausing movement. During this phase raw data from vehicle-mounted sensors including odometric sensors and rangefinders are logged to a file throughout teaching for later processing.

Abstract: A guidance, navigation, and control method and system for an underground mining vehicle wherein an operator directly or indirectly guides the vehicle along a route as vehicle-mounted sensors including odometric sensors and rangefinders log motion data to a file which data is processed into a route profile including a vehicle path, a sequence of submaps along the path, and a profile of desired speed as a function of distance along the path. Thereafter the vehicle automatically repeats the route as represented by the route profile without an operator using the sensors and the maps to determine vehicle location.

Abstract: The present invention provides a system for coordinating multiple vehicles in a passageway environment (e.g., in underground mines). The system includes methods and apparatus for determining the global position and orientation of a vehicle in said passageway environment, and methods for planning routes and monitoring the travels of multiple vehicles in said passageway environment. A global position and orientation estimation system employs one or more odometric sensors and one or more range sensing devices. It works in three basic steps. In the first step, it records and processes sensor data that is descriptive of the passageway environment by moving the system through said passageway environment. In the second step, it generates a globally consistent map of said passageway environment.

Abstract: The present invention provides a guidance, navigation, and control method and system for an underground mining vehicle that allow said vehicle to be taught a route by a human operator and then have it automatically drive the route with no human intervention. The method works in three steps: teaching, route profiling, and playback. In the teaching step the vehicle is manually driven by a operator (or using tele-operation whereby the operator views a screen displaying live views from vehicle-mounted cameras and using remote controls) along a route which can consist of an arbitrary sequence of maneuvers including tramming forwards, switching directions, tramming backwards, turning, or pausing movement. During this phase raw data from vehicle-mounted sensors including odometric sensors and rangefinders are logged to a file throughout teaching for later processing.