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

Vehicle Controls & Behaviors

Annual Plan

Novel Data-Driven Algorithms for Autonomous Vehicle Path Planning Problems with Uncertain Data Parameters

Project Team

Principal Investigator

Saravanan Venkatachalam, Wayne State University

Government

Jonathon Smereka, Sam Kassoumeh, Calvin Cheung, U.S. Army GVSC

Industry

Scott Corey, Timothy Bail, Spatial Integrated Systems (SIS)

Student

Mark Crawford, Venkata Siri Chirala, Nur Banu Demir, Wayne State University

Project Summary

Project began 2019.

During the exploratory phase and first year of the project, the focus was to develop offline high-level planning algorithms for a team of ground vehicles while considering system and environment uncertainties, leader-follower pattern with unmanned ground vehicles (UGVs) following the manned leader. The current phase of the project focuses on two stages following the planning stage, and are as follows: evaluation stage – the data on vehicle casualties, time spent on point of interests (POIs), and travel time are continuously collected from the field, and the feasibility and quality of the solution obtained from planning stage is continuously monitored; and execution stage – obtain quick recourse actions for the teams whenever the current solution loses feasibility or deteriorates in quality or a new ad-hoc task gets added. For example, a convoying mission will employ manned ground vehicles and UGVs in a leader-follower pattern with several UGVs following the single manned leader. Beyond the offline plans, if some of the assets are compromised or fail in a hostile attack or more time is required at certain POIs or if any a new POI is added then what are the quick recourse solutions that will help ensure the team can complete the mission?

This project focuses on the following fundamental question that arises during the evaluation and execution stages of deploying a team of GVs: How to estimate the sensitivity of the solutions obtained during offline planning, and what are the quick recourse (online) solutions in the event of deviations in the resources availability and environments?

Publications:

  • S. Venkatachalam, M. Bansal, J.M. Smereka, J. Lee, “Two-stage stochastic programming approach for path planning problems under travel time and availability uncertainties,” https://arxiv.org/abs/1910.04251
  • “Stochastic Programming Model for Unmanned Ground Vehicles” GVSETS, Autonomous Ground Systems Technical Session, 8/13/19.

From Prior Work closely related to the proposed project:

  1. S. Faridimehr, S. Venkatachalam, R. Chinnam. “A stochastic programming approach for electric vehicle charging network design,” IEEE Trans. on Intelligent Transportation Systems, 99 (2018): 1-13.
  2. K. Sundar, S. Venkatachalam, S. Rathinam. “Analysis of mixed-integer linear programming formulations for a fuel-constrained multiple vehicle routing problem,” Unmanned Systems, 5.04 (2017): 197-207.
  3. K. Sundar, S. Venkatachalam, S. Rathinam. “Formulations and algorithms for the multiple depot, fuel- constrained, multiple vehicle routing problem,” IEEE American Control Conference (ACC), 2016.
  4. K. Sundar, S. Venkatachalam, S. Manyam. “Path planning for multiple heterogeneous Unmanned Vehicles with uncertain service times,” IEEE Intern. Conf. on Unmanned Aircraft Systems, 2017.
  5. M. Bansal, K.L. Huang, S. Mehrotra. “Decomposition algorithms for two-stage distributionally robust mixed binary programs,” SIAM Journal on Optimization, 28.3: 2360-2383, 2018.
  6. M. Bansal, K.L. Huang, S. Mehrotra. “Tight second stage formulations in two-stage stochastic mixed integer programs,” SIAM Journal on Optimization, 28.1: 788-819, 2018.

#1.30