High-accuracy Dead-reckoning System (HADRS) for Manned and Unmanned Ground Vehicles

Transcription

1 Mobile Robotics Lab High-accuracy Dead-reckoning System (HADRS) for Manned and Unmanned Ground Vehicles PI: Johann Borenstein* Research Professor at the University of Michigan * 28 years experience in GPS-free position estimation for mobile robots and pedestrians

2 High-accuracy Dead-reckoning System (HADRS) HADRS works on any wheeled or tracked ground vehicle. Works on any terrain, including rugged, steeply inclined terrain. Requires only a power connection and wheel/track encoder data from the host vehicle, or our proprietary, retrofittable add-on encoders. Optional components: retrofittable add-on encoders can be mounted to almost any vehicle in less than one hour no need to access vehicle s own odometry system Trajectory-plotting OCU for tele-operator support. HADRS installed on a Packbot. HADRS fits neatly into 1/3 of the payload bay of a Packbot, but can be mounted on any vehicle

3 Features Works without GPS, computer vision, beacons, or any other external references but can accommodated and integrate optional external GPS. Is indifferent to environment: indoors, outdoors, urban, or rural. Is small-sized (fits into 1/3 Packbot payload bay). Is designed for minimal power consumption. Outputs JAUS compatible, time-stamped X, Y, Z, and heading data stream at 10 Hz. Outputs data via Ethernet, WiFi, RF, or long-range data modem. Is unaffected by extreme operating temperatures.

4 Performance For missions of up to 30 minutes duration: average position error: 0.5% of distance traveled even on rugged terrain; maximum position error: 1% of distance traveled Average heading error: 1 Errors may be larger on deformable terrain (sand, mud) or any other terrain that promotes excessive track slippage. For longer missions Position errors increase at ~1% for every 30 minutes of operation.

5 HADR Description Main Components Controlled by a Gumstix computer-on-module. Sensor modalities: NavChip inertial measurement unit (IMU) KVH single-axis fiber optic gyro (FOG) Host s odometry OR our add-on encoder system. Active temperature stabilization for FOG and IMU. Optional: Integration with external GPS receiver. Operator console unit Version 1: Text only Version 2: trajectory plotter to support tele-operation HADR System IMU Closed-loop temperature stabilization Optional OCU (For convenience only, not needed at runtime) Optional WiFi, USB, or RF modem Initialization, mode selection, debugging Computed pose Streaming at 10 Hz Gumstix Computer-on-Module USB Hub KVH FOG Closed-loop temperature stabilization Ethernet or WiFi Add-on encoders (optional) Host Computer External GPS (optional) Main components of the HADRS

6 Retrofittable Add-on Encoders Add-on encoder sub-system comprises two encoders. Each encoder consists of 1. a pair of latching hall-effect sensors mounted to the robot s chassis next to a track sprocket or wheel, 2. a laser-cut plastic encoder disk sized to fit inside the sprocket Function: Each encoder disk has between 4 to 18 press-fit neodymium magnets arranged in alternating north-south orientation. As the sprocket turns, alternating magnetic poles pass over the hall-effect sensors, generating quadrature signals. Retrofittable add-on encoder system, here fitted to a Packbot track drive sprocket

7 Y [m] Experimental Results with HADRS Run 1 Red curve: Faint orange curve: Total travel distance: Total travel time: Trajectory produced by HADRS Trajectory from consumer-grade GPS used for ground truth. (GPS got a fix only at Point A, therefore the GPS trajectory starts only at A. ) 976 m 23.5 min Return Position Error: 1.5 m (= 0.15% of total distance) Largest position error anywhere during this run (near x = -190 m, y = -270 m): ~3 m (= 0.3% of total distance) Steepest slope traversed: ~17 on grass (see photo inset taken near Point A) Challenges in this run: partially sloping terrain, zig-zagging south of A, traverse of several curbs. X [m] A Trajectory of Run 1 with the HADR system installed on a Packbot To visualize the slope of the terrain, the photo inset HADRS Position shows the experimenter standing near Point A -270

8 Y [m] Experimental Results with HADRS Run 2 Red curve: trajectory recorded by the HADR system Total travel distance: Total travel time: 676 m 28.5 min Return Position Error: 1.04 m (=0.15% of total distance traveled) Largest position error anywhere during this run: ~2 m (=0.3% of total distance traveled) Steepest slope traversed: on grass Challenges in this experiment: partially sloping terrain, traverse of several curbs 10 X [m] HADRS Position Slope down ~10-15 o Trajectory of Run 2 with the HADR system installed on a PackBot

9 Acknowledgements and Disclaimers Acknowledgements Parts of this Project Agreement Holder (PAH) effort was sponsored by the U.S. Government under Other Transaction number W 15QKN between the Robotics Technology Consortium, Inc, and the Government. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon. Some parts of this effort were supported by the Ground Robotics Reliability Center (GRRC) at the University of Michigan, with funding from government contract DoD-DoA W56H2V through the Joint Center for Robotics. Disclaimer The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the U.S. Government.