Posted on

Andy’s Status Report for September 27

At the beginning of this week, I looked deeper into drone design and found that the requirements were more complex than expected. Beyond the standard components, we also needed ESCs, a flight controller, a gyroscope, and an altimeter. These essential parts alone would consume over half of our budget, and adding a USB camera and a high-resolution thermal camera would push us well beyond our limits. Because of this, we decided to pivot away from drones.

I then explored the idea of indoor search-and-rescue drones. This use case was more reasonable since limiting operations to indoor spaces removes the challenge of long-range control. However, I still faced difficulties with the RC control aspects of drone design.

After meeting with our instructors, we formally decided to switch from drones to indoor ground vehicles. I began researching suitable platforms and initially considered the iRobot Create, but since it is no longer widely available, I shifted focus to on-shelf robot car kits, such as the ELEGOO UNO. While these kits are basic, they provide a reliable chassis we can expand upon without losing time on building vehicles from scratch. This means that we can put more time on algorithm designs, which we are are better at.

Next, I will focus on how to handle vehicle rotation and how to integrate lidar into the system. Although the project change has set me slightly behind schedule, I am confident I can quickly catch up with this more practical and achievable direction.

Posted on

Soren’s Status Report for September 27

This week I continued learning about and working on algorithms and techniques for detecting people using IR imaging data, as well as looked into which specific hardware components (cameras and controllers) would be best suited for our project. Significant time this week has also gone into the design review presentation and slides.

Because we have made a pivot in the project (based on the fact that our original idea of using a drone to search an area for people would not have been able to cover a wide enough area to make for an effective product) we have pivoted to a different project idea, but one that will still be making use of detecting people using IR imaging, so this component of the project will still be useful and is not one that we are behind schedule in. In the next week I hope to have finally placed an order for an IR camera based on what I have found out about them this week, and test if simpler algorithms (such as edge detecting, or thresholding) will be sufficient to detect people (and not detect not-people) for our use case, or if more advanced methods (CNNs, for instance) will be needed.

Posted on

Team Status Report for September 27

Our design has been significantly changed since last week as a result of difficulties with our previous design that led us to modify our use case. The main difficulties we encountered before were the cost of using multiple drones and that an affordable thermal camera would not be able to cover a large enough area with a high enough resolution. We did not think there was a way to address these difficulties while meeting the use case requirements and providing a project that would be helpful for our use case because our drone would not be able to cover enough area quickly while maintaining reasonably high detection accuracy.

Our new use case is to build a search and rescue robot for indoor use after an event such as a gas leak. This contains similar technical challenges of detecting people and of doing pathing, but avoids the problems we could not solve with our previous use case. We are still working on the design, and are currently planning on using a circular robot so that it can rotate easily, with 2D SLAM (Simultaneous Localization and Mapping) with Lidar to construct the environment to be used for pathing, and a thermal camera to detect people in the environment, but the design could change as we finalize it in the coming days.

The main risk given our change in use case is that we still do not have a complete design at this point. This is especially difficult as the design must be completed in time for the design review presentation on Monday. However this deadline does help to manage the risk in that we will be completing our design quickly in order to be prepared for the presentation. Our schedule has changed as a result of the change to our project, but has not yet been finalized as we are still working on the design. An updated schedule will be provided in our design review presentation.

Posted on

Jeremy’s Status Report for September 27

This week I worked on the design for the path planning aspect of the project. However, since our design pivoted towards the end of the week, my previous design for the path planning is no longer particularly useful, so I had to begin looking at entirely different path planning starting Friday the 25th.

My prior path planning algorithm design was for multiple drones to search a large, outdoor area. This design involved first partitioning the search area into equal-area sections that could be assigned to drones, and then plotting a path for each drone to its section that could sweep it out using a simple rectangular pattern. A slide I had made for the design presentation on this algorithm is shown below:

 

However, since we are moving towards having a robot on the ground explore an indoor building, this path planning algorithm is not applicable to our new use case because the search area will have far more obstacles and be in an unknown configuration. So, I have begun looking into Simultaneous Localization and Mapping (SLAM) and using pathing algorithms with that. The pathing will now have to be done onboard the robot rather than programmed beforehand, since the environment will be unknown at the start and determined in real time using SLAM. Based on some initial research, I am thinking that we will want to use Lidar for detecting walls in the environment.

The change in design has led me to be behind our original schedule since I had previously wanted to have the path planning fully designed and start working on implementation at this point. To catch up I plan to spend a lot of time over the weekend on designing the SLAM and new pathing, although with our design changes we will have to develop a new schedule for the project. By next week I hope to have fully designed the SLAM and path planning algorithm for our current project.

Posted on

Soren’s Status Report for September 20

The main thing I worked on this week was learning about and implementing algorithms to take thermal imaging data and tell where sources of heat in a given image are. For the most part I am on schedule, because I should be able to work on putting a thermal image processing algorithm on a Raspberry Pi next week as outlined in our planned schedule for the project, however there will likely need to be quite a bit more adjustments made to our thermal image processing algorithm (in terms of kind of data being received, and possible optimizations). I think this can be accomplished in the coming week and week after.

Posted on

Andy’s Status Report for September 20

This week, I assisted Jeremy in completing the proposal presentation. My primary focus has been on designing a testing drone, and I aim to finish this work by the middle of next week. Currently, I am slightly behind schedule due to other obligations. To address this, I plan to dedicate more time to the project next week to get back on track.

Posted on

Team Status Report for September 20

One significant risk for the project is that, in addition to the complexity of the individual systems, integrating the systems into a working product will also be difficult. To manage this risk we plan to coordinate well and complete the individual subsystems quickly to allow a lot of time for integration.

Another significant risk is that we need to find a way to do test flights of the drone. To manage this risk we plan on contacting testing sites and applying for use of them early, as soon as next week even though the drone won’t be ready to fly yet just to make sure we have access to a way to do test-flights.

Early on in the week, there was significant design development, refinement, and modification as we were working on the project proposal presentation. There have not been any further changes to the design of the system yet, and the current design can be found in the project proposal slides.

Posted on

Jeremy’s Status Report for September 20

The main thing I did this week was work on the project proposal presentation. Early on in the week I worked a lot on developing the slides for the project proposal, and then I also did additional preparation since I was the one giving the presentation.

Later on in the week, I also did some preliminary research on drone path planning algorithms. My progress is on schedule at the moment. By the end of next week I hope to have completed development of an initial version of the path planning algorithm we will use to program our drones.

Posted on

Introduction and Project Summary

Our project is to build a drone surveying system for disaster relief. This system would be used for search and rescue after a natural disaster. The group of drones would be assigned an area of land that needs to be checked for people and report back the locations of people found. Because we plan on using multiple drones to increase the area of land that can be checked for people in need of help in a given period of time, we will have to develop algorithms to plot out the optimal paths for each drone and ensure that they are checking mostly independent areas. We plan to have a thermal imaging camera on each drone as well as a normal camera to allow the drones to detect people. Then we will also need to develop an image processing algorithm to detect people from the data.