Control Issues of an Autonomous Vehicle

Abstract

Abstract-This paper addresses some control issues of a robotic amphibious vehicle that can serve as a general framework for automation of tractors used in construction. These include the vehicle's low-level dynamic equations, the development of its braking control system, kinematics in interactions with ground and the slip problem. Simulation and real-time results to date are presented.Index Terms-Unmanned Ground Vehicle, dynamic and kinematic modelling, skid-steering, sliding mode control, slip I. INTRODUCTIONHE ARGO is an ongoing important project using a robotic amphibious vehicle as the experimental platform, currently developed at the ARC Centre of Excellence for Autonomous Systems (CAS).The vehicle is a fully autonomous outdoor one that can serve as a general framework for automation of tractors used in construction. The robotic vehicle is automated from a 20hp, 3m x 1.45m x 1.1m, 0.5 ton automotive amphibious vehicle that can achieve 30km/h on land and 3km/h on water. The vehicle is equipped with throttle and left/right actuators, angular speed encoders, GPS, and other sensors for control and navigation.The vehicle, shown in Fig. 1, uses a V-twin combustion engine. Power is transmitted from the engine to the eight wheels through a continuous variable transmission (CVT), gearbox, differential and a chain system. The vehicle wheels are connected by a chain system and driven by the left and right outputs of a differential. Two brake discs are attached to the outputs of the differential and can be operated separately. The differential and braking system enables turning the vehicle (skid-steering). The driveline of the vehicle, shown in Fig. 2, consists of the engine, CVT, gearbox, differential (in gearbox), chain system, and eight wheels.

A compact design of zero-radius steering autonomous amphibious vehicle with direct differential directional drive - UTAR-AAV

“…A servo and two DC motors were used to power the UTAR-AAV, while the obstacles avoidance system is fed by input signals such as those from visual processing sensors, rotary encoders, and GPS. An ongoing ARGO robotic amphibious vehicle [7] research list control issue when automating the vehicle as one of the major challenges faced, especially braking during vehicle steering, the relationship between the hydraulic pressure on brake discs and the piston position remains highly nonlinear, making it difficult to model and control the vehicle's turning and steering. This vehicle consists of an engine, CVT gearbox, differential (in gearbox), chain system, and eight wheels and may not tap into the advantages provided by an electric vehicle.…”

Position control of a 6×6 ATV using a MIMO fuzzy controller

“…Derivations of the dynamic model of a similar "Argo" model can be found in [8] and is continued in [9]. The interaction between vehicle and terrain is explored extensively in [10], which itself is an extension of preliminary work done in [9].…”

“…The interaction between vehicle and terrain is explored extensively in [10], which itself is an extension of preliminary work done in [9]. Unfortunately, the models are highly parameterized.…”

Integrated planning and control of large tracked vehicles in open terrain

“…It takes waypoint lists, which are coordinates, as its inputs from the path planner and generates speed and yaw rate. The trajectory tracking control is inspired by various sources, which include the Argo [13] and the RedRover [14]. The latter can be traced back to the control model developed by Kanayama [15].…”