Browsing by Author "Tsoeu, Mohohlo S."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Open Access
    Rollover prevention and path following of a scaled autonomous vehicle using nonlinear model predictive control
    (2018) Gwayi, Isaac; Tsoeu, Mohohlo S.
    Vehicle safety remains an important topic in the automotive industry due to the large number of vehicle accidents each year. One of the causes of vehicle accidents is due to vehicle instability phenomena. Vehicle instability can occur due to unexpected road profile changes, during full braking, obstacle avoidance or severe manoeuvring. Three main instability phenomena can be distinguished: the yaw-rate instability, the rollover and the jack-knife phenomenon. The main goal of this study is to develop a yaw-rate and rollover stability controller of an Autonomous Scaled Ground Vehicle (ASGV) using Nonlinear Model Predictive Control (NMPC). Open Source Software (OSS) known as Automatic Control and Dynamic Optimisation (ACADO) is used to design and simulate the NMPC controller based on an eight Degree of Freedom (8 DOF) nonlinear vehicle model with Pacejka tire model. Vehicle stability limit were determined using load transfer ratio (LTR). Double lane change (DLC) steering manoeuvres were used to calculate the LTR. The simulation results show that the designed NMPC controller is able to track a given trajectory while preventing the vehicle from rolling over and spinning out by respecting given constraints. A maximum trajectory tracking error of 0.1 meters (on average) is reported. To test robustness of the designed NMPC controller to model mismatch, four simulation scenarios are done. Simulation results show that the controller is robust to model mismatch. To test disturbance rejection capability of the controller, two simulations are performed, with pulse disturbances of 0.02 radians and 0.05 radians. Simulations results show that the controller is able to reject the 0.02 radians disturbance. The controller is not able to reject the 0.05 radians disturbance.
  • Thumbnail Image
    Item
    Open Access
    Stereo visual simultaneous localisation and mapping for an outdoor wheeled robot: a front-end study
    (2019) Wolf, Ryan Evan; Tsoeu, Mohohlo S.
    For many mobile robotic systems, navigating an environment is a crucial step in autonomy and Visual Simultaneous Localisation and Mapping (vSLAM) has seen increased effective usage in this capacity. However, vSLAM is strongly dependent on the context in which it is applied, often using heuristic and special cases to provide efficiency and robustness. It is thus crucial to identify the important parameters and factors regarding a particular context as this heavily influences the necessary algorithms, processes, and hardware required for the best results. In this body of work, a generic front-end stereo vSLAM pipeline is tested in the context of a small-scale outdoor wheeled robot that occupies less than 1m3 of volume. The scale of the vehicle constrained the available processing power, Field Of View (FOV), actuation systems, and image distortions present. A dataset was collected with a custom platform that consisted of a Point Grey Bumblebee (Discontinued) stereo camera and Nvidia Jetson TK1 processor. A stereo front-end feature tracking framework was described and evaluated both in simulation and experimentally where appropriate. It was found that scale adversely affected lighting conditions, FOV, baseline, and processing power available, all crucial factors to improve upon. The stereo constraint was effective for robustness criteria, but ineffective in terms of processing power and metric reconstruction. An overall absolute odometer error of 0.25-3m was produced on the dataset but was unable to run in real-time.