CAG*-EDP: Chaos-enhanced Ant-Genetic, A* Algorithm hybrid with Error Detection and Patching

Primary Investigator (PI) Name

Muhammad Hassan Tanveer

Department

SPCEET - Robotics and Mechatronics Engineering

Abstract

The works within this project aim to introduce a fine navigation system that accounts for both path planning and gait efficiency for quadrupedal robots. A chaotic ant-genetic-A* algorithm (CAG*) is developed to enrich the will-be-combined individual benefits of the A* Algorithm, Ant Colony Optimization (ACO), and a chaos-enhanced Genetic Algorithm (GA) whilst minimizing their detriments, a process to be done through an overhead monocular camera and motion capture which serve as stand-ins for UAVs and GPS data, respectively. Error Detection and Patching (EDP) will call for the correction of the Unitree Go1 Dog’s (the main unit of experimentation) gait sequence in the event it goes astray and/or adopts an inefficient walk cycle when moving autonomously, in tandem with enforcing an efficient, minimalist approach for path planning by only using CAG* to develop routes as needed rather than continuously.

Disciplines

Computer Sciences | Robotics

This document is currently not available here.

Share

COinS
 

CAG*-EDP: Chaos-enhanced Ant-Genetic, A* Algorithm hybrid with Error Detection and Patching

The works within this project aim to introduce a fine navigation system that accounts for both path planning and gait efficiency for quadrupedal robots. A chaotic ant-genetic-A* algorithm (CAG*) is developed to enrich the will-be-combined individual benefits of the A* Algorithm, Ant Colony Optimization (ACO), and a chaos-enhanced Genetic Algorithm (GA) whilst minimizing their detriments, a process to be done through an overhead monocular camera and motion capture which serve as stand-ins for UAVs and GPS data, respectively. Error Detection and Patching (EDP) will call for the correction of the Unitree Go1 Dog’s (the main unit of experimentation) gait sequence in the event it goes astray and/or adopts an inefficient walk cycle when moving autonomously, in tandem with enforcing an efficient, minimalist approach for path planning by only using CAG* to develop routes as needed rather than continuously.