Project Topics 2012-13

Robot Navigation

Autonomous Explorations, Inc., and WPI are working on a Navigation system for an autonomous robot. We are looking for a student to work with us on a summer internship or MQP. The work includes writing programs to perform data collection and analysis, getting the hardware to interface properly, calibration, and performing a range of systems tests.

 We cannot offer any money (yet, but keep reading). We can offer an exciting opportunity to help us develop and test a system that might someday go to the Moon or Mars. Or get turned into a commercial product. And write it up, perhaps to present at a conference. And if our proposal to NASA, which was approved but not funded, ever gets funded, then we could pay you, too, if/when you are a graduate student.

Advisor: Mike Gennert

Tunneling Robot

Burying utility cable under roads involves digging through the road, laying the cable, and then repairing the road surface. Better would be to build a gopher-bot to tunnel under roads to lay cable. The project team will design, build, and evaluate a tunneling robot. The robot will be remote-controlled from the ground.

Advisors: Mike Gennert, William Michalson

Kayak Robot

Prof. Chris Kitts and his students at Santa Clara University in California have been building robotic kayaks for maritime applications. The project team will design and build a kayak robot. We will then evaluate it against Santa Clara robots at the annual Tahoe-Palooza at Lake Tahoe. Let’s show them what real Robotics Engineers can do!

Advisors: Michael Gennert, Taskin Padir

Tree-Climbing Robot

The project team will continue development of a tree-climbing robot. The robot will be remote-controlled from the ground and will be used for tree inspection. We learned a lot from the 1st MQP. Most important: It is all about legs and actuation!

Advisors: Michael Gennert, William Michalson

Curvic Qualification

Blisks (bladed disks) are used in jet engines and consist of a central hub surrounded by curved blades. The manufacturing of a blisk usually requires numerous inspection steps. One of these steps is to determine whether an assembly mating feature called the 'curvic' (a toothed area at the top of the central hub) is concentric to other part features. Another is to determine whether the end faces of the blisk are parallel. Specialized jigs and tools as well as trained inspection operators are used to do this at the present time. The expected outcome is a robotic process that can automatically perform these quality checks. Some of the challenges in developing such a process involve:

  •  Preparing and aligning the jig,
  •  Mounting and aligning the blisk in the jig,
  • Making the measurements, and
  • Marking the blisk to indicate areas out of tolerance.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Blisk Visual Defect Inspection

Blisks (bladed disks) are used in jet engines and consist of a central hub surrounded by curved blades. The manufacturing of a blisk usually requires numerous inspection steps, one of which is that the entire surface area of the part must be visually inspected for nicks, dings, scratches and other surface anomalies. The expected outcome is a robotic process that can automatically find suspect areas and bring them to the attention of manufacturing personnel. Some of the challenges in developing such a process involve:

  • Determining how to identify defects (or perhaps by inverting the problem, determining areas that do not have any identifiable problems), and
  • Alerting the operators to suspect areas.

 While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Deburring Outer Bands

Static (non-rotating) jet engine compressor stages are typically assemblies of 'sectors'. A sector is an angular segment of the disk before assembly. It consists of an 'inner band' with (typically) curved blades radiating outward to an 'outer band'. The machining of the outer band segment often leaves metal burrs on various of the machined edges. The expected outcome is the development of a robotic process that can automatically find and then remove the burrs. Some of the challenges in developing such a process involve:

  • Determining the presence of burrs, and
  • Removing the burrs without damaging the part.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Automated Tool Prep/Tool Crib

Jet engine components are often made from materials such as Titanium and Inconel. The manufacturing processes often require machining away a considerable amount of metal. Given the hardness of the metals involved, the process wears out end mills at a rapid rate. The expected outcome is the development of a robotic process that can automatically change end mills in the tool holders. Some of the challenges in developing such a process involve:

  • Positioning the robot to the tool holder,
  • Opening the tool holder and removing the used end mill, * Inserting a new end mill into the holder to the proper length, and
  • Tightening (closing) the tool holder.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Tool Tracking

A variety of machine tools are used in the manufacturing of jet engine components. Worn or misaligned tools can lead to manufacturing quality control problems. The expected outcome of this project is a process that can automatically track where various tools are being used in the production facility, and then feed that information into a manufacturing process quality control system so that use of particular tools can be correlated with the production of substandard parts. Some of the challenges in developing such a process involve:

  • Uniquely identifying the tools (RFID tags are one possibility),
  • Noting where and when a tool is being used (machine, job, etc.), and
  • Feeding this information into the existing manufacturing process quality control systems.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Tube Cutoff & End Finishing

A large number of tubes of various shapes, diameters and lengths need to be created as part of the manufacturing of a gas turbine engine. The expected outcome is a robotic process that can automatically cut off / trim the end of a tube and then prepare the end of the tube for the next manufacturing step. Some of the challenges in developing such a process involve:

  • There are many (thousands) of tubes to be handled and the number is steadily increasing. The system needs to be able to deal with this level of complexity.
  • The tubes have often have been bent into various shapes and sizes. Again, the system needs to be able to deal with this level of complexity.
  • Deburr & polish the end of the tube after cutting.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Tube Welding

A large number of tubes of various shapes, diameters and lengths need to be created as part of the manufacturing of a gas turbine engine. As part of the manufacturing of these tubes, various types of flanges are inserted onto the ends of the tubes which are then welded in place by an orbital welder. The expected output is a robotic process that can automatically insert the flange onto the end of a tube and then move the tube into a jig for the orbital welding step. Some of the challenges in developing such a process involve:

  • There are many (thousands) of tubes to be handled and the number is steadily increasing. The system needs to be able to deal with this level of complexity.
  • Inserting the flange onto the end of the tube without damaging the flange or the tube.
  • Performing the weld.
  • Inspecting the weld once it has been completed.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Automatic Jig Construction

A large number of tubes of various shapes, diameters and lengths need to be created as part of the manufacturing of a gas turbine engine. As part of the manufacturing process, the tubes finished tube assemblies are loaded into specially built jigs to determine if they have been bent and assembled correctly. The jigs are often large in size, so they cannot be left intact due to floor space restrictions. Thus they are disassembled and then reassembled when needed again. The expected output is a robotic process that can automatically assemble and qualify the testing jigs. Some of the challenges in developing such a process involve:

  • There are many (thousands) of tubes to be handled and the number is steadily increasing. The system needs to be able to deal with this level of complexity.
  • The jigs need to be created on demand
  • The jigs need to be built to a high engineering tolerance.
  • The jigs should ideally be created out of a supply of common / standardized component parts.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Tube Polishing

A large number of tubes of various shapes, diameters and lengths need to be created as part of the manufacturing of a gas turbine engine. As part of the manufacturing of these tubes, various types of flanges are inserted onto the ends of the tubes which are then welded in place by an orbital welder. The expected outcome is a robotic process that can automatically polish out the discolorations that result from the welding step. Some of the challenges in developing such a process involve:

  • There are many (thousands) of tubes to be handled and the number is steadily increasing. The system needs to be able to deal with this level of complexity.
  • Positioning the welded tube assembly for the polishing operation on each weld.
  • Performing the polishing without damaging the tube assembly.
  • Inspecting the polished area to determine if it has been polished enough.
  • Loading and unloading the tube assemblies from the feed stations.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Wear Sleeve Prep

A large number of tubes of various shapes, diameters and lengths need to be created as part of the manufacturing of a gas turbine engine. As part of the manufacturing of these tubes, various locations along the tube are given a 'wear sleeve' to protect the tubing. The expected outcome is a robotic process that can automatically apply the wear sleeve materials. These consist of specially pre-cut pieces of a carbon fiber material along with special tapes. In the manual process presently used, the tape is dispensed from a motorized dispenser which cuts the tape pieces to length automatically. Some of the challenges in developing such a process involve:

  • There are many (thousands) of tubes to be handled and the number is steadily increasing. The system needs to be able to deal with this level of complexity.
  • Positioning the wear sleeve and tape onto the tube and applying it (wrapping it around the tube).
  • Loading and unloading the tube assemblies from the feed stations.

While it is unlikely to be the case, U.S. Citizenship may be required for project participants.

Advisors: Craig Putnam, Stephen Nestinger

Blade Root Quality Check

Blisks (bladed disks) are used in jet engines and consist of a central hub surrounded by curved blades. The manufacturing of a blisk usually requires several steps, one of which is to check the quality of the filet at the root of the blade. The expected outcome is a robotic process that can automatically perform this quality check. Some of the challenges in developing such a process involve:

  • Determining whether there is a better way to accomplish the task than the current manual process, and
  • Avoiding causing damage to the blisk during the inspection.

While it is unlikely to be the case, U.S. Citizenship may be required for project participant.

Advisors: Craig Putnam, Stephen Nestinger

Blisk Overspray Removal

Blisks (bladed disks) are used in jet engines and consist of a central hub surrounded by curved blades. The manufacturing of a blisk usually requires several coating steps, one of which results in overspray onto certain areas of the blisk. The overspray is currently being removed by a labor-intensive manual process using rubberized abrasive materials (Cratex), large 'Q-Tip' type devices and very fine metal files. The expected outcome of the project is a robotic process that could automatically find and remove the overspray. Some of the challenges in developing such a process involve:

  • Determining the areas of overspray
  • Removing the overspray, and
  • Avoiding causing damage to prior coatings under the oversprayed areas.

While it is unlikely to be the case, U.S. citizenship may be required for project participants.

Advisors: Stephen Nestinger and Craig Putnam

Stand-Alone Sensor and Actuator Modules for Robot Applications in Healthcare

The project team will design and realize stand-alone sensor (encoder, inertial measurement, ranging) and actuator (servo, stepper) modules that can easily be integrated with systems and devices used in healthcare (wheelchairs, walkers, hospital beds). It is an expected outcome of the project that the completed modules will be demonstrated on a semi-autonomous mobile robot. This project is partially sponsored by the National Science Foundation, as a result, the project team will closely work with Professor Padir and the graduate students in the Robotics and Intelligent Vehicles Research (RIVeR) Lab.

Advisor: Taskin Padir

Development of an Autonomous Blimp

I. Project Description

The purpose of this project is to design and build an autonomous, outdoor, lighter-than-air vehicle for the demonstration of autonomous navigation functions and use in multiple roles though interchangeable hardware packages.

II. Current State of Project

The current iteration of the project has designed and built a hardware platform consisting of a gondola containing all of the motor control hardware, sensor systems, and power systems, as well as space and connectors for a mission module. Hardware has also been created to support a range extension hardware package and a “drop module” capable of lifting and dropping small packages. There is currently code for a basic autopilot and hardware interface. The autopilot will direct the blimp to a set of preprogrammed waypoints; however it does not compensate in real time for wind or follow terrain.

III. Potential follow-up projects

There are several areas for follow on projects to improve on the current blimp system which can be addressed individually or combined into a larger project. The first of these is to improve the hardware of the blimp. The current design is very modular, and switching out individual components with better replacements can be done without affecting the entire system. There is also potential for several projects relating to the software of the blimp, most of which have to do with improving the capabilities of the autopilot or adding both software and hardware to increase the blimp’s capabilities, for example improving navigation and adding obstacle avoidance. Finally, there are many potential projects related to creating new hardware packages. Hardware packages can give the blimp new abilities, allowing it to perform new roles.

IV. Recommended Follow-up projects

The project that we feel would best refine the current platform would be one that incorporates the refinement of the power system, the general improvement of the autopilot, and the design, construction, and programming of one or two hardware packages. Alternatively, a project to give the blimp the capability to avoid objects would also be viable. This would involve the installation of new sensors, the improvement of the blimp’s maneuvering capabilities, and the creation of new software.

If you have any questions about these potential projects, the current team can be reached at blimpmpq@wpi.edu or see our website

Advisor: Fred Looft

 
  • Email a Friend
  • Bookmark this Page
  • Share this Page