Parikshit Maini

Parikshit Maini

Auto-generated profile, based on publicly available information

Parikshit Maini

Department of Computer Science and Engineering, University of Minnesota, Minneapolis, MN, USA

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 May 2019 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
PMParikshit Maini
Parikshit Maini
KSKaarthik Sundar
Kaarthik Sundar
MSMandeep Singh
Mandeep Singh
SRSivakumar Rathinam
Sivakumar Rathinam
PSP.B Sujit
P.B Sujit
Low cost unmanned aerial vehicles (UAVs) need multiple refuels to accomplish large area coverage. We propose the use of a mobile Ground Vehicle (GV), constrained to travel on a given road network, as a refueling station for the UAV. Determining optimal routes for a UAV and GV, and selecting rendezvous locations for refueling to minimize coverage time is NP-hard. We develop a two-stage strategy for coupled route planning for UAV and GV to perform a coverage mission. The first stage computes refueling sites that ensure reachability of all points of interest by the UAV and feasible routes for both the UAV and GV. In the second stage, Mixed-Integer Linear Programming (MILP)-based exact methods are developed to plan optimal routes for the UAV and GV. As the problem is NP-Hard, we also develop computationally efficient heuristics that can find good feasible solutions within a given time limit. Extensive simulations are conducted to corroborate the effectiveness of the developed approaches. Field experiments are also performed to verify the performance of the UAV-GV solution.

24 February 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
PMParikshit Maini
Parikshit Maini
BGBurak M. Gonultas
Burak M. Gonultas
VIVolkan Isler
Volkan Isler
Thearea used for grazing cattle constitutes about one-third of the land in United States. In this paper, we present the design of Cowbot, an autonomous weed mowing robot for maintaining cow pastures. Cowbot is designed to operate in rugged environments and provides a cost-effective method for weed control in cow pastures. Path planning for the Cowbot is challenging since weed distribution on pastures is unknown. Given an onboard sensor with a limited field of view, online planning is necessary to detect weeds and plan paths to mow them. We study the general online coverage planning problem for an autonomous mower with curvature and field of view constraints. We develop two algorithms that are able to utilize new information on weed detection to optimize path length and ensure coverage. We deploy our algorithms on the Cowbot and perform field experiments to validate the suitability of our methods for real-time path planning. We also perform extensive simulation experiments which show that our algorithms result in up to 60% reduction in path length as compared to boustrophedon and random-search based coverage paths.

24 August 2020 in IEEE Transactions on Automation Science and Engineering
Verified
Journal Article
Closed access
PMParikshit Maini
Parikshit Maini
PTPratap Tokekar
Pratap Tokekar
PSP.B Sujit
P.B Sujit
Persistent monitoring on terrains using mobile robotic sensors requires coordinated planning. Terrain features add visibility obstacles and limited fuel capacity of aerial robots leads to range restrictions that make the problem challenging. We address the visual-monitoring problem on piecewise linear features within a terrain using multiple mobile robots for persistent operations. The planner must account for visual coverage, refueling aerial robots during the mission, and placement of refueling depots while also utilizing the available sensor diversity to minimize overall costs for the monitoring mission. Building on previous works on visibility in specific classes of polygons and fuel-constrained routing, we develop a discrete representation of the problem that allows the design and application of discrete optimization techniques to find optimal solutions. We develop a mixed-integer linear programming (MILP) formulation and discuss a branch-and-cut implementation to compute exact solutions. We also develop a construction heuristic based on the idea of competitive construction of robot paths using a step-increment strategy. We report the results from computational simulations and illustrate proof of concept using experiments on real robots. Note to Practitioners —This article is motivated by the need to perform persistent monitoring in applications, such as border patrol and perimeter surveillance. Unmanned aerial and ground robots can be used to perform these activities uninterruptedly. However, aerial robots have limited fuel capacity and need periodic refueling. Hence, the number of refueling depots and their placement within the environment also affects the monitoring task. Also, due to terrain variation, robots are subject to limited visibility. Therefore, we need to consider refueling constraints and terrain visibility aspects while planning optimal routes for the robots to perform visual monitoring. In this article, we present a general optimal routing formulation to compute exact solutions. We also present a fast heuristic for real-time applications that produce feasible solutions. The algorithms are validated in simulations. We also show a proof of concept using experiments in limited outdoor settings.

21 May 2021 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
PMParikshit Maini
Parikshit Maini
PTPratap Tokekar
Pratap Tokekar
PSP.B Sujit
P.B Sujit
We study the problem of visually monitoring a set of points on a 2.5D terrain using an unmanned aerial vehicle (UAV) with a downward-facing camera. The goal is to find a tour of minimum length for the UAV offline to visually inspect all points of interest. Varying terrain and limited field of view of the camera restrict the visibility of the UAV and can create obstacles in the flight path, making the problem challenging. The problem is NP-hard and generalizes the traveling salesperson problem (TSP). We present several algorithms to solve this problem. Our main theoretical contribution is a constant-factor approximation algorithm (assuming fixed parameters for the terrain). We also present a practical algorithm that uses the solution to a generalized TSP (GTSP) subinstance. We benchmark the GTSP-based algorithm using a branch-and-cut integer linear programming formulation and find that the proposed algorithm scales to much larger instances and is computationally fast. We also show proof-of-concept using field deployment of a UAV to visually monitor points of interest in the environment.