{"id":{"repo_id":"stellenbosch","oai_identifier":"oai:scholar.sun.ac.za:10019.1/136134"},"canonical_url":"https://search.dev.ndltd.org/etd/stellenbosch/oai:scholar.sun.ac.za:10019.1/136134","repository":{"repo_id":"stellenbosch","name":"Stellenbosch University","base_url":"https://scholar.sun.ac.za/server/oai/request"},"display":{"title":"Cooperative Collision Avoidance for the Voyager Unmanned Ground Vehicle: Practical Implementation and Demonstration","abstract":"The following thesis presents the research work done in practically implementing a cooperative collision avoidance system on the Voyager mobile robotic platform provided by the CSIR. The project adapts approaches proposed by Hughes for unmanned aerial vehicles and adapts and practically implements them on the physical Voyager robot. The system’s motion planning was split into a global planner, focused on finding a long-term route between the vehicle’s start position and a given goal position though a previously mapped environment of static obstacles, and a local planner that can predict and cooperatively resolve short-term collisions with static obstacles or other vehicles en route. The global planner uses the A* graph search algorithm to find an optimal route through a roadmap made of the Voronoi diagram of the map. The local planners of cooperative agents take a decentralised approach, broadcasting short-term predictions of their future positions to each other over a shared communication channel. If an agent detects a conflict between its own predicted trajectory and the predicted trajectory of any of the other vehicles, a structured exchange of messages is initiated according to which agents establish a priority order and sequentially plan local evasive trajectories. Evasive plans are found by using Dijkstra’s graph search algorithms on a tree of prescribed manoeuvres to find a conflict free manoeuvre sequence that minimises the agent’s deviation from the globally planned motion. A Stanley controller, augmented with an in-track proportional controller, is used to execute the planned trajectories, along with additional logic for switching between following the long-term reference trajectory and the short-term collision avoidance trajectory. The system was practically tested on the physical Voyager robot in a real-world indoor environment. Illustrative multi-vehicle conflict scenarios were tested using one physical vehicle and a number of virtual vehicles. The experimental results show that the vehicles can: successfully plan and follow long-term routes through the environment, predict en route collisions between vehicles or with the environment, as well as cooperatively plan and execute short-term collision avoidance trajectories.","abstract_html":"The following thesis presents the research work done in practically implementing a cooperative collision avoidance system on the Voyager mobile robotic platform provided by the CSIR. The project adapts approaches proposed by Hughes for unmanned aerial vehicles and adapts and practically implements them on the physical Voyager robot. The system’s motion planning was split into a global planner, focused on finding a long-term route between the vehicle’s start position and a given goal position though a previously mapped environment of static obstacles, and a local planner that can predict and cooperatively resolve short-term collisions with static obstacles or other vehicles en route. The global planner uses the A* graph search algorithm to find an optimal route through a roadmap made of the Voronoi diagram of the map. The local planners of cooperative agents take a decentralised approach, broadcasting short-term predictions of their future positions to each other over a shared communication channel. If an agent detects a conflict between its own predicted trajectory and the predicted trajectory of any of the other vehicles, a structured exchange of messages is initiated according to which agents establish a priority order and sequentially plan local evasive trajectories. Evasive plans are found by using Dijkstra’s graph search algorithms on a tree of prescribed manoeuvres to find a conflict free manoeuvre sequence that minimises the agent’s deviation from the globally planned motion. A Stanley controller, augmented with an in-track proportional controller, is used to execute the planned trajectories, along with additional logic for switching between following the long-term reference trajectory and the short-term collision avoidance trajectory. The system was practically tested on the physical Voyager robot in a real-world indoor environment. Illustrative multi-vehicle conflict scenarios were tested using one physical vehicle and a number of virtual vehicles. The experimental results show that the vehicles can: successfully plan and follow long-term routes through the environment, predict en route collisions between vehicles or with the environment, as well as cooperatively plan and execute short-term collision avoidance trajectories.","abstract_has_math":false,"creators":["Mtimkulu, Thando"],"institution":"Stellenbosch : Stellenbosch University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Engelbrecht, Jacobus Adriaan Albertus"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026-03","date_published":"2026-03","updated_at":"2026-07-24T04:40:06Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholar.sun.ac.za/handle/10019.1/136134","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Engelbrecht, Jacobus Adriaan Albertus"]},{"key":"dc:contributor.other","label":"Dc Contributor Other","values":["Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering."]},{"key":"dc:creator","label":"Author","values":["Mtimkulu, Thando"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-04-23T08:11:20Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-04-23T08:11:20Z"]},{"key":"dc:date.issued","label":"Date","values":["2026-03"]},{"key":"dc:publisher","label":"Institution","values":["Stellenbosch : Stellenbosch University"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://scholar.sun.ac.za/handle/10019.1/136134"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (MEng)--Stellenbosch University, 2026.","Mtimkulu, T. 2026. Cooperative Collision Avoidance for the Voyager Unmanned Ground Vehicle: Practical Implementation and Demonstration. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/2ed1d8ea-b7eb-422e-8ef2-faa82597bcf8"]},{"key":"dc:description.abstract","label":"Abstract","values":["The following thesis presents the research work done in practically implementing a cooperative collision avoidance system on the Voyager mobile robotic platform provided by the CSIR. The project adapts approaches proposed by Hughes for unmanned aerial vehicles and adapts and practically implements them on the physical Voyager robot. The system’s motion planning was split into a global planner, focused on finding a long-term route between the vehicle’s start position and a given goal position though a previously mapped environment of static obstacles, and a local planner that can predict and cooperatively resolve short-term collisions with static obstacles or other vehicles en route. The global planner uses the A* graph search algorithm to find an optimal route through a roadmap made of the Voronoi diagram of the map. The local planners of cooperative agents take a decentralised approach, broadcasting short-term predictions of their future positions to each other over a shared communication channel. If an agent detects a conflict between its own predicted trajectory and the predicted trajectory of any of the other vehicles, a structured exchange of messages is initiated according to which agents establish a priority order and sequentially plan local evasive trajectories. Evasive plans are found by using Dijkstra’s graph search algorithms on a tree of prescribed manoeuvres to find a conflict free manoeuvre sequence that minimises the agent’s deviation from the globally planned motion. A Stanley controller, augmented with an in-track proportional controller, is used to execute the planned trajectories, along with additional logic for switching between following the long-term reference trajectory and the short-term collision avoidance trajectory. The system was practically tested on the physical Voyager robot in a real-world indoor environment. Illustrative multi-vehicle conflict scenarios were tested using one physical vehicle and a number of virtual vehicles. The experimental results show that the vehicles can: successfully plan and follow long-term routes through the environment, predict en route collisions between vehicles or with the environment, as well as cooperatively plan and execute short-term collision avoidance trajectories."]},{"key":"dc:title","label":"Title","values":["Cooperative Collision Avoidance for the Voyager Unmanned Ground Vehicle: Practical Implementation and Demonstration"]}]}],"canonical_facts":{"dc:contributor.advisor":["Engelbrecht, Jacobus Adriaan Albertus"],"dc:contributor.other":["Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering."],"dc:creator":["Mtimkulu, Thando"],"dc:date.accessioned":["2026-04-23T08:11:20Z"],"dc:date.available":["2026-04-23T08:11:20Z"],"dc:date.issued":["2026-03"],"dc:description":["Thesis (MEng)--Stellenbosch University, 2026.","Mtimkulu, T. 2026. Cooperative Collision Avoidance for the Voyager Unmanned Ground Vehicle: Practical Implementation and Demonstration. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/2ed1d8ea-b7eb-422e-8ef2-faa82597bcf8"],"dc:description.abstract":["The following thesis presents the research work done in practically implementing a cooperative collision avoidance system on the Voyager mobile robotic platform provided by the CSIR. The project adapts approaches proposed by Hughes for unmanned aerial vehicles and adapts and practically implements them on the physical Voyager robot. The system’s motion planning was split into a global planner, focused on finding a long-term route between the vehicle’s start position and a given goal position though a previously mapped environment of static obstacles, and a local planner that can predict and cooperatively resolve short-term collisions with static obstacles or other vehicles en route. The global planner uses the A* graph search algorithm to find an optimal route through a roadmap made of the Voronoi diagram of the map. The local planners of cooperative agents take a decentralised approach, broadcasting short-term predictions of their future positions to each other over a shared communication channel. If an agent detects a conflict between its own predicted trajectory and the predicted trajectory of any of the other vehicles, a structured exchange of messages is initiated according to which agents establish a priority order and sequentially plan local evasive trajectories. Evasive plans are found by using Dijkstra’s graph search algorithms on a tree of prescribed manoeuvres to find a conflict free manoeuvre sequence that minimises the agent’s deviation from the globally planned motion. A Stanley controller, augmented with an in-track proportional controller, is used to execute the planned trajectories, along with additional logic for switching between following the long-term reference trajectory and the short-term collision avoidance trajectory. The system was practically tested on the physical Voyager robot in a real-world indoor environment. Illustrative multi-vehicle conflict scenarios were tested using one physical vehicle and a number of virtual vehicles. The experimental results show that the vehicles can: successfully plan and follow long-term routes through the environment, predict en route collisions between vehicles or with the environment, as well as cooperatively plan and execute short-term collision avoidance trajectories."],"dc:identifier.uri":["https://scholar.sun.ac.za/handle/10019.1/136134"],"dc:language.iso":["en"],"dc:publisher":["Stellenbosch : Stellenbosch University"],"dc:title":["Cooperative Collision Avoidance for the Voyager Unmanned Ground Vehicle: Practical Implementation and Demonstration"],"dc:type":["Thesis"]},"updated_at":"2026-07-24T04:40:06Z"}