Changes

Several of the RobotX team members in late 2019, designed and constructed a CrawlerBot (see http://ocrobotx.org/mediawiki/index.php/Holonomic_Robotic_Platform#Project_Purpose) using holonomic drive techniques in anticipation of using that experiential knowledge as an approach vector to design a more complex WAM-V propulsion system.  It is apparent to the experienced team members that a combination of straightforward linear propulsion systems that can be mechanically articulated into a holonomic drive system presents the best of both options: speed when needed, and fine-grained maneuverability required to complete complex navigation maneuvers.

Several of the RobotX team members in late 2019, designed and constructed a CrawlerBot (see http://ocrobotx.org/mediawiki/index.php/Holonomic_Robotic_Platform#Project_Purpose) using holonomic drive techniques in anticipation of using that experiential knowledge as an approach vector to design a more complex WAM-V propulsion system.  It is apparent to the experienced team members that a combination of straightforward linear propulsion systems that can be mechanically articulated into a holonomic drive system presents the best of both options: speed when needed, and fine-grained maneuverability required to complete complex navigation maneuvers.

Power management presents a large opportunity for design consideration with respect to the computing network and infrastructure for sensor data processing, and AI backbone.  Raspberry PI 4 computing has proven itself to be a very capable processing platform, hosting Python, MATLAB APIs, and other AI plugins.  Distributed computing of multiple PIs can provide cluster computing performance, at literally a fraction of the super-computer cost, and using a miniscule amount of the power consumption. Linux, ROS, Python, and MATLAB are the primary software staples of the processing infrastructure required to complete the project, as well as embedded microcontroller devices similar to Arduinos.

Power management presents a large opportunity for design consideration with respect to the computing network and infrastructure for sensor data processing, and AI backbone.  Raspberry PI 4 computing has proven itself to be a very capable processing platform, hosting Python, MATLAB APIs, and other AI plugins.  Distributed computing of multiple PIs can provide cluster computing performance, at literally a fraction of the super-computer cost, and using a miniscule amount of the power consumption. Linux, ROS, Python, and MATLAB are the primary software staples of the processing infrastructure required to complete the project, as well as embedded microcontroller devices similar to Arduinos.

===First Order Sensor Development===

====First Order Sensor Development====

Three primary sensor inputs, as well as advanced processing must be achieved to meet minimum competition status, namely:  LIDAR, Vision and perception, and GPS/IMU. The EE/CS teams will divide the tasks among smaller working groups with articulated development and test events, as well as key subsystem integration milestones.  LIDAR processing, vision mapping and AI perception, as well as occupancy grid development becomes critical as the project timeline progresses, since all of the competition tasks rely on mapping the field and interacting appropriately with the objects located in the range.  The team has allocated funds to construct training aids to be deployed along with the WAM-V test phases, providing a competition range facsimile for training.

Three primary sensor inputs, as well as advanced processing must be achieved to meet minimum competition status, namely:  LIDAR, Vision and perception, and GPS/IMU. The EE/CS teams will divide the tasks among smaller working groups with articulated development and test events, as well as key subsystem integration milestones.  LIDAR processing, vision mapping and AI perception, as well as occupancy grid development becomes critical as the project timeline progresses, since all of the competition tasks rely on mapping the field and interacting appropriately with the objects located in the range.  The team has allocated funds to construct training aids to be deployed along with the WAM-V test phases, providing a competition range facsimile for training.

 

===Second Order Sensor Development===

===Second Order Sensor Development===

Secondary sensor development involves depth sounding/echo sounding sensors for bottom contour mapping which facilitates ground collision avoidance. Finally, SONAR ping detection, discrimination, ranging, and localization is also required to negotiate several of the competition challenges.

Secondary sensor development involves depth sounding/echo sounding sensors for bottom contour mapping which facilitates ground collision avoidance. Finally, SONAR ping detection, discrimination, ranging, and localization is also required to negotiate several of the competition challenges.