This research project focused on two specific work related hazards of pipe installation, cave-ins and underground utilities. Both thrusts addressed the problems through technological interventions. This final report presents an exciting and successful endeavor that produced tangible results ready to be deployed in the field to save lives and increase productivity at the same time. The following will briefly discuss the most significant accomplishments. Even though diverse support systems such as shoring, shielding, and sloping are to be applied during trench excavating and pipe laying operation, accidents still occur. The underlying concept for finding a remedy for this problem is to remove the need for people to enter the confined space through the use of a tele-robotic approach. In response to the many complexities in installing pipes of various types, two separate mechanisms were designed, fabricated, and field tested. While both were invented as attachments to backhoe excavators or cranes, the first device, labeled PipeMan, is capable of handling large and heavy concrete pipes used to build drainage systems. The second, called PipeMan Jr, is tailored to manipulate lighter but longer water or sewer pipes built with O-ring gasket joints also called push-on-joint type seals. PipeMan and PipeMan Jr each required a uniquely different technical approach to create a safe alternative to the present hands-on method. Tele-robotic systems are mechanical devices that combine human and machine intelligence to perform a task remotely with the assistance of various sensors, computers, man-machine interface devices, and electronic controls. The last generation of PipeMan utilized a fork and clamp system leading to a field assessed cycle time for laying one pipe segment of 3.6 min. Moreover, a crew of 3 is able to perform pipe installation instead of the conventional 5, which drastically increases productivity. It is estimated that Pipeman would be to lay 70 pieces of 8 ft (2.4 m) pipes per day if laid consecutively into an open trench. Because of the requirement to apply significant normal forces for jointing pipes with O-ring gaskets the simple "push" approach of PipeMan was ineffective. PipeMan Jr draws on four key enabling technologies: 1) A hydraulically powered pipe-jointer, 2) compliant pipe carrier, 3) two powered struts to adjust the position of the pipe-end in x an z directions, and 4) wireless control interface for the operator. PipeMan Jr. was also repeatedly field tested with actual equipment operators performing the tasks. Currently trenching contractors have to rely on color-coded markings applied by utility locators to know where utilities are buried in the ground. As accidents prove time and time again, this method is unreliable. The second main thrust of this project was to study the effectiveness of non-intrusive sensing systems to detect and locate buried utilities of all types, especially nonmetallic plastic pipes. Furthermore, the final technology should be equipment-mountable and function like a fish-finder. The Buried-Utility-Detection-System (BUDS) uses a multi-sensory approach to underground sensing. It is believed that more reliable results are possible because it decreases the degree of ambiguity and increases the probability. One example is the integration of the Electro-Magnetic- Induction (EMI) with spatial data from an actuator carrying the sensing antenna termed the Equipment-Mounted BUDS (EM-BUDS.) Software was programmed to control the manipulator while, in parallel, collecting and processing sensory data. A unique human-machine interface uses the manipulator as a "pointing stick" that tells the operator not only that there is a buried pipe but also where its approximate location is. This system was tested in the field. Principal Investigator/Program Director: Bernold, Leonhard, E. 2 PHS 398/2590 (Rev. 05/01) Page Continuation Format Page Extensive experiments were conducted with Ground-Penetrating-Radar (GPR) technology since it promised to detect even non-metallic objects. Combined with the EMI and mounted on a mobile platform it was indeed able to distinguish between buried concrete with and without steel rebar. The most critical progress in our ability to detect underground utilities was made in improving the effectiveness of processing GPR data. By using a wavelet based approach to filtering raw data from the GPR sensor we were able to make drastic advancements in pinpointing the location of buried pipes made of different materials. A large pipe-laying project in connection with a new Chilled-Water facility on NC State's campus provided the testing ground for many of the technologies developed for this project. The most dramatic opportunity was to compare our utility sensing results with the actual situation found during trench excavation. It is anticipated that the analysis of the collected data will continue for years to come due to the fact that so much of it was collected during the extensive pipe-laying project.