Showing 1–2 of 2 results for author: Venkataram, H S
-
Scheduling the NASA Deep Space Network with Deep Reinforcement Learning
Authors:
Edwin Goh,
Hamsa Shwetha Venkataram,
Mark Hoffmann,
Mark Johnston,
Brian Wilson
Abstract:
With three complexes spread evenly across the Earth, NASA's Deep Space Network (DSN) is the primary means of communications as well as a significant scientific instrument for dozens of active missions around the world. A rapidly rising number of spacecraft and increasingly complex scientific instruments with higher bandwidth requirements have resulted in demand that exceeds the network's capacity…
▽ More
With three complexes spread evenly across the Earth, NASA's Deep Space Network (DSN) is the primary means of communications as well as a significant scientific instrument for dozens of active missions around the world. A rapidly rising number of spacecraft and increasingly complex scientific instruments with higher bandwidth requirements have resulted in demand that exceeds the network's capacity across its 12 antennae. The existing DSN scheduling process operates on a rolling weekly basis and is time-consuming; for a given week, generation of the final baseline schedule of spacecraft tracking passes takes roughly 5 months from the initial requirements submission deadline, with several weeks of peer-to-peer negotiations in between. This paper proposes a deep reinforcement learning (RL) approach to generate candidate DSN schedules from mission requests and spacecraft ephemeris data with demonstrated capability to address real-world operational constraints. A deep RL agent is developed that takes mission requests for a given week as input, and interacts with a DSN scheduling environment to allocate tracks such that its reward signal is maximized. A comparison is made between an agent trained using Proximal Policy Optimization and its random, untrained counterpart. The results represent a proof-of-concept that, given a well-shaped reward signal, a deep RL agent can learn the complex heuristics used by experts to schedule the DSN. A trained agent can potentially be used to generate candidate schedules to bootstrap the scheduling process and thus reduce the turnaround cycle for DSN scheduling.
△ Less
Submitted 9 February, 2021;
originally announced February 2021.
-
'healthWISE' An Android Application For Personal Health And Nutrition Management
Authors:
Hamsa Shwetha Venkataram,
Kruthika,
Prarthana MK,
Shivani C,
Siva Naga Suresh Purama,
Muthukumar SP
Abstract:
Now-a-days, people are getting more health conscious and tend to keep a check on nutritional gain from the packed food products they consume. This application christened 'healthWISE' helps a mobile user to scan/enter the bar-code on the packed food product, know the nutritional information, and upon entering the quantity to be consumed, it displays the energy the individual can consume. If the use…
▽ More
Now-a-days, people are getting more health conscious and tend to keep a check on nutritional gain from the packed food products they consume. This application christened 'healthWISE' helps a mobile user to scan/enter the bar-code on the packed food product, know the nutritional information, and upon entering the quantity to be consumed, it displays the energy the individual can consume. If the user has exceeded the stipulated amount of calories, then it suggests the exercise to burn the extra number of calories.
△ Less
Submitted 7 December, 2015;
originally announced December 2015.
