Impact on knowledge and technique:
- We are successfully building a custom drone network. We have completed test flights to validate the robustness of our drones’ platform and evaluate the performance of our ML algorithms.
- We have validated high resolution environmental sensor ready to be deployed at TFA (techforall.org), located in the heart of one of Houston’s most economically disadvantaged neighborhood and near petro-chemical refineries. Moreover, our drones are in their 3rd generation and are extremely stable in flight and ready to perform longer range missions.
- We have deployed a FROG system for measurement of VOCs (specifically targeting benzene, toluene, and xylene (or BTEX) compounds). Future ground-based VOC data will be used to determine when to launch drone-based sampling, and hence this work contributes to development of methodologies for data-driven mobile missions.
- The developed efficient machine learning adaptation and continuous learning techniques is one critical progress for us to achieve the goal of on-drone real-world data-driving mobile sensing missions.
- Our designed inference algorithms and application-aware drone navigation methods allow us to overcome the current limitations of drone-based environmental sensing (that is due to drone instability and vibrations) and hence ensure a high effectiveness in assisting first responders during extreme events.
- Our novel machine-learning algorithms enable efficient on-drone sensing and learning and networked drone mission planning strategies to provide effective navigation strategies, potentially benefiting first responders in addressing extreme events.
- Our study underscores the significance of accurate AoA estimation in enhancing performance and offers insights for the development of more efficient massive MIMO systems for drone networks.
- Our study underscores the significance of accurate AoA estimation in enhancing performance and offers insights for the development of more efficient massive MIMO systems for drone networks.
Impact on other disciplines:
- Our designed QEPAS sensor will allow future drone environmental applications to benefit from an unprecedented high-resolution real-time sensing.
Impact on physical resources that form infrastructure:
- This project has allowed an enhancement of the instrumentation available to Rice investigators to monitor air quality.
- New sensing probes that measure multiple gases in addition to VOC have been purchased and integrated to the current pollution aerial sensing platform.
Impact on development of human resources:
- Our educational activities have included, post-doctoral scholar, graduate student, and undergraduate student training, and conference, industry, and workshop presentations.
- Our educational summer internship program allows our students to strengthen their research and engineering skills and be ready to start for some of them a PhD program in the coming fall.
- Summer internship programs that we offer in our labs provide valuable research experience to undergraduates and help them better transition to Ph.D. programs if they decide to pursue graduate degree.
Impact on institutional resources that form infrastructure:
- New equipment to measure BTX compounds was purchased, learned, and deployed. This instrument can potentially be used in future air quality studies.
Impact on technology transfer:
- Rice university is world leader in QEPAS technology and strongly collaborate with the THORLABS joint industry-university research lab PolySense. Recently THORLABS have released an acoustic detection module based on the PolySense designs.In collaboration with PolySense Rice has provided:
- the first demonstration of QEPAS sensors capable to detect gas outdoor for several days and on mobile systems
- the hyperspectral nature of QEPAS by simultaneously operating a near-IR and a mid-IR laser to contemporary detect two gas species
- the multi-gas QEPAS technology for mobile and airborne operation.
Impact on society beyond science and technology:
- Our developed efficient machine learning adaptation and continuous learning techniques pave the way for more extensive applications of machine learning empowered intelligent functionality in resource-constrained IoT devices.
- The portion of the project team doing BTX measurements at TFA is interacting with local staff, allowing for increased education regarding local air quality issues.
- The market for environmental monitoring market is estimated to grow from $18.4 billion in 2019 to $25.5 billion in 2024. Rising pollution levels, favorable regulatory scenario, ongoing installation of environmental monitoring stations, development of environment-friendly industries, increasing awareness and legislation on pollution monitoring, expansion of pollution monitoring infrastructure, (especially as emerging economies such as China, India, Mexico and others become increasingly climate and environmentally conscious), are the major market drivers. The air pollution monitoring segment and particulate detection segment account for the largest shares of the environmental monitoring market, by component and by application, respectively.
- Our collected gas concentration data is being regularly uploaded to our community mobile app, which as a result helps our community learn about how the air quality is affecting their health and how they should adapt their lifestyle to reduce the pollution harmful effects.
- Our designed mobile app and regularly uploaded gas concentration data enable the local community to monitor the air quality in their neighborhood and take necessary steps to reduce the harmful effects of pollution.