DigSafe Autonomous Cable Detection

Morris, Leo; Holden, Grace; Heffern, Victoria

Student Work

DigSafe Autonomous Cable Detection

Public Deposited

Underground power cables must be marked by utility providers before any groundbreaking work can be performed on a work site. Currently, technicians mark cables by hand, which can be time-consuming, monotonous, and dangerous. The Dig Safe Major Qualifying Project aims to develop a prototype robot to perform autonomous underground electrical cable detection and marking. The robot will navigate a work site, detect buried cables, and accurately mark them in compliance with Massachusetts Law. Based upon the previous work executed by the two teams from years prior, our goals were to strengthen the subsystems - cable detecting, following, and marking. Within these subsystems, we sought to identify the areas that need improvements, determine how they can be improved, and conclude how these changes can affect the efficiency of the whole system. To this extent, several key discoveries were made, such as the motion controller sending inconsistent signals, spray arm breaking due to its weight, and the camera being unable to retrieve data from the wand. Through our development stages - we mitigated the sensor interferences, refined the wand data retrieval, reconstructed the spray arm and its motion planning, and developed a new cable prediction algorithm. Additionally, we provided further developments to the navigation stack, allowing for further testing to be executed autonomously.

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

Subject

Publisher