Skip to main content
arc logo
Automotive Research Center
hero image
Back to all projects

Vehicle Controls & Behaviors

Annual Plan

A Decision-Based Mobility Model for Semi and Fully Autonomous Vehicles

Project Team

Principal Investigator

Vijitashwa Pandey, Oakland University

Government

David Gorsich, U.S. Army GVSC

Industry

Judson Estes, Fiat Chrysler

Student

Christopher Slon (postdoc), Line Deschenes, Sam Kassoumeh (SMART fellow), Oakland University

Project Summary

Project began in 2018 and was completed in 2021.

utility theoretic method

The objective of this work is to develop a decision-theoretic method to define and measure the mobility for ground vehicles exhibiting partial to full-autonomy. More generally, this research provides a rigorous methodology of evaluating ground vehicle systems and thereby help with acquisition, operational and sustainment decisions pertaining to them. Some example decision scenarios where the method provides guidance are:

  • A given ground vehicle platform is robust but also cost prohibitive to sustain.
  • A path planned through a hostile terrain may be quicker, but jeopardizes vehicle safety.
  • Should additional terrain strength information be acquired, given the cost and time constraints?

A software tool for decision making is developed. The tool begins with the capturing of decision maker’s preferences in the form of a utility function based on the responses to statistically generated questions. While the multilinear utility function is an option, we have developed techniques to generalize them to include binary attributes, ones that have a cut-off level, and also when the attribute set is large. A value-function based assessment method has also been incorporated which uses regression/ML to first encode tradeoff behavior. The tool uses a probability wheel based assessment for capturing preferences under risk, which helps with convergence to an acceptable value. For a set of alternatives and associated uncertainties (acquired through tests or simulations), the tool returns the best alternative, the relative attribute-wise performance gap between pairs of alternatives, and value of information on stochastic alternatives.

Publications:

  • Pandey, V., Bos, J., Ewing, J., Kysar, S. et al., “Decision-Making for Autonomous Mobility Using Remotely Sensed Terrain Parameters in Off-Road Environments,” SAE Int. J. Adv. & Curr. Prac. in Mobility 3(4):1682-1689, 2021, https://doi.org/10.4271/2021-01-0233.
  • Mollan, C, Pandey, V, Slon, C, & Gorsich, D. “Sequentially Utility Maximizing Path Planning Using a Distributed Pool Architecture.” Proceedings of the ASME 2021 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. Volume 3B: 47th Design Automation Conference (DAC). Virtual, Online. August 17–19, 2021. V03BT03A008. ASME. https://doi.org/10.1115/DETC2021-67946
  • Pandey, V., Slon, C., Deschenes, L., Gorsich, D., & Jayakumar, P. (2020). A Decision Based Mobility Model for Semi and Fully Autonomous Vehicles. SAE International Journal of Advances and Current Practices in Mobility, 2(2020-01-0747), 2640-2649.
  • Kassoumeh, S., Majcher, M., Ealy, J., Gorsich, D., Jayakumar, P., & Pandey, V. (2020). Balancing lifecycle sustainment cost with value of information during design phase. SAE International Journal of Advances and Current Practices in Mobility, 2(2020-01-0176), 2451-2458.
  • Pandey, V., Slon, C., Mollan, C., Barthlow, D., Gorsich, D., and Jayakumar, P., 2020, “Utility Function Derived Off-Road Vehicle Path Planning”, Proceedings of the 2020 ASME International Design Engineering Technical Conferences, St. Louis MO (moved virtual).
  • Barthlow, D., Pandey, V., Gorsich, D., and Jayakumar, P., 2020, “Off-Road Vehicle Path Planning Using Geodesics on a Multifactor Terrain Model”, Proceedings of the 2020 ASME International Design Engineering Technical Conferences, St. Louis MO (moved virtual).
  • Slon, C., Pandey, V., Gorsich, D. and Jayakumar, P., 2020. “Reconciling Simultaneous Evolution of Ground Vehicle Capabilities and Operator Preferences” SAE World Congress, Detroit MI. Technical Paper 2020-01- 0172.

#1.25