Software and Hardware Implementation for the Akoya and Bandit Satellites

Forrest Rogers-Marcovitz
Advisers: Dr. Arye Nehorai, Ed Richter, Dr. Michael Swartwout

Department of Electrical And Systems Engineering
Washington University in St. Louis
Spring 2008


This semester capped three years of work on the Akoya and Bandit satellites. For the majority of that time, I have acted as the Embedded Systems Lead - responsible for flight software and electronics. My time this semester was spent on a number of systems through out the satellites. In the power system, I complied code to read both analog-to-digital (A2D) senors and Dallas 1-wire sensors; in addition, calibration constants were determined for all A2D sensors. Telemetry frames were created for both power and ADC systems; these Telemetry frames are grabbed and parsed by the ground station. In addition, I oversaw redesigns of the Bandit and dock electronics. This work was presented at the Nanosat-5 Critical Design Review, April 21, 2008.

I. Introduction

The goal of the Bandit project is to demonstrate remote operation of an under-5kg service vehicle. This includes repeatable docking and stationkeeping (keeping the vehicle in an assigned orbit) with a host spacecraft as well as blended autonomous control (having the ability to be operated manually or autonomously). In the future we intend to demonstrate the feasibility of using a university Nanosat-class spacecraft and Bandit-class service vehicle for more advanced missions.

Akoya-B is a 25kg student-built spacecraft, designed to operate in low-Earth orbit as the parent vehicle for the Bandit mission. Akoya can also be used as the host satellite for generic payloads, such as biological experiments. It was developed in collaboration with Santa Clara University, and demonstrates rapid integration and testing of flight hardware.

The two satellites are being developed for the University Nanosat Program as run by the Air Force Research Laboratory, NASA, and American Institute of Aeronautics and Astronautics (AIAA). Select schools participate in a 24-month satellite design competition; the winning school delivers a flight-ready spacecraft for a future, secondary launch.

In this report, I present my major work this semester. Section II describes the power sensors, and Section III describes telemetry gathering. I describe the changes to the dock support electronics in Section IV and conclude in Section V.


II. Power System Sensors

The power system contains analog sensors that monitor system's state including: battery temperature, solar panel input, and power output (voltage and current). These sensors are feed to four analog-to-digital (A2D) converter chips which talk to the power system via a Serial Peripheral Interface (SPI) bus. The A2D chips have a 10-bit precision with a range of zero to five volt input. The power systems acts as the SPI master and polls these devices regularly; I finished and verified the supporting software. Along with Anne Schneider, we tested all the sensors and determined calibrated constans to convert from the 10-bit output to meaningful sensor values such has temperature.

The power system also acts as the Dallas Bus Master. Dallas 1-Wire is a device communications bus system designed by Dallas Semiconductor that provides low-speed data, signaling and power over a single wire, plus a ground wire. Each system board has four Dallas 1-wire devices including a temperature sensor and switches that control the latch-up modules and turn systems on and off. Using code from Santa Clara University, I was able to poll all the sensors on the Engineering Design Unit and control the power state of the subsystems. In the past, capacitance problems have occurred on flight satellite, thus limiting the number of sensors found. The flight satellite is undergoing a wire harness configuration change to combine data and power buses; so testing will wait until that is complete.

III. Telemetry Gathering

Each subsystem compiles its own telemetry frame which consists of sensors and state variables. Different telemetry frames are made for different operating modes depending on the frequency and detail need for each state variable. Right now you must query each subsystem individually for its telemetry packet. In the future these packets will automatically be complied by the commsys and stored on the power's flash memory for future downlink. While the ground station is within communication range of the satellite, the telemetry frames will automatically sent to the ground station. The telemetry frames consisted of predetermined byte structure as defined by the telemetry frames spreadsheet.

Telemetry Command

You can send this command to any subsystem to request a telemetry packet. Each frame is broken up into packets determined by a fixed number of bytes defined by the code. You send the frame type and page number you want - start counting with 0. The first response is for a successful send, in which the telemetry packet is returned. The first byte of this packet is the source EDP address, the second byte is the frame type, the third byte is page number followed by the total number of pages for the frame, the fifth byte is the number of bytes of returned parameters (not counting header info) and the remaining bytes are the parameters of the telemetry packet. The second response is if the ID of the requested frame does not exist. If this happens, the scheduler will return an error.

The telemetry packets for the power and ADC (attitude determination and control) systems have be compiled, and the ground station can poll and parse these frames. Further work includes expanding the telemetry frames to include the reaming systems and tailoring the frame types for different operation modes.

IV. Bandit and Dock Electronics

The dock support board consists of all electronics that support the Bandit mission and that live on the Akoya spacecraft. The main requirements are as follows:

  • Dock Control and Sensing
  • Akoya-Bandit Communication
  • Image Capture
  • Image Downlink

The old dock support board ncludes two 8-bit Atmel microprocessors. One is dedicated to dock control and Bandit command communication (417 MHz); this Atmel also runs the elevator and door motors, solenoid pins, and controls Bandit’s charging. The second microprocessor is in charge of image capture and downlink. Images from Bandit (900 MHz) and the dock are sent as analog video and digitalized by the frame capture board; this microprocessor then downlinks the images and performs LED finding. Both microprocessors are linked to Akoya’s data bus through I2C and the Emerald data protocol.

Old Dock Support Board

Three set of changes were made.The first set of changes were needed to match changes made to Bandit’s electronics. Bandit was updated to include a more powerful processor and provide better communications. This involves eliminating the 900 MHz and 417 MHz cross-communication radios along with their support electronics. These radios are being replaced by radios supporting the Zigbee protocol that will handle both command and image traffic. Bandit will be able to perform image navigation without support on the dock support board. The second set of changes increase the dock’s sensing abilities. This involves adding charge and motor current sensing, switch sensors, and the full set of encoder outputs.The third set of changes improve image processing. We plan on accomplishing this by eliminating the frame capture board and digitalizing the video.

These requirements where met by replacing the 8-bit image processor with a more powerful 32-bit processor. This creates a cleaner design that eliminates the Frame Capture Board and mirrors the image processing done on Bandit. The dock’s cameras will remain the same, and a chip will be added to convert the analog NTSC video signal in to a digital VGA signal. The updated processor can handle image processing and encoding of a digital signal.

New Dock Support Board

The schematics and board layout for the Dock Support Board and Bandit Electronics were created by Lane Haury and Keith Swaback. I was responsible for verifying that the schematics meet all design requirements and everything was connected properly. The Dock Support board consists of the main board plus a smaller daughter board that contains the AVR32 processor and associated electronics such as RAM. The Bandit Electronics consists of a power board, a sensor board and a command board with the same daughter board as the dock support board. Three are the board have been sent out for manufacturing and the remaining two should be sent out within the week.

V. Conclusion

Much progress was made this semester on the satellite electronics and software towards the flight system. The new dock and bandit electronics should be the last major design change in the satellite. The major task that remains is the software for and testing of the new electronics which will be preformed over the summer. The remaining systems have only minor changes needed; a complete task lists will be completed before my graduation. My remaining time here will be spent finishing up as many loose ends as possible and making sure that all my systems are well documented for the new software and electronics team.


VI. Relevant Works