Two-layer obstacle collision avoidance with machine learning for more energy-efficient unmanned aircraft trajectories

Choi, Youngjun; Jiménez, Hernando; Mavris, Dimitri N.

doi:10.1016/j.robot.2017.09.004

Cited by 43 publications

(21 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The simulation results provide good insights into the latency performance of the exchange of data, reliability, and latency reduction in the update cycle. In [35], the authors proposed an algorithm for obstacle avoidance based on two layers. The global path optimization layer gives information about the obstacles from sensors using a clustering technique.…”

Section: Related Workmentioning

confidence: 99%

Traffic Flow Management of Autonomous Vehicles Using Platooning and Collision Avoidance Strategies

et al. 2021

View full text Add to dashboard Cite

Connected Autonomous Vehicles (AVs) promise innovative solutions for traffic flow management, especially for congestion mitigation. Vehicle-to-Vehicle (V2V) communication depends on wireless technology where vehicles can communicate with each other about obstacles and make cooperative strategies to avoid these obstacles. Vehicle-to-Infrastructure (V2I) also helps vehicles to make use of infrastructural components to navigate through different paths. This paper proposes an approach based on swarm intelligence for the formation and evolution of platoons to maintain traffic flow during congestion and collision avoidance practices using V2V and V2I communications. In this paper, we present a two level approach to improve traffic flow of AVs. At the first level, we reduce the congestion by forming platoons and study how platooning helps vehicles deal with congestion or obstacles in uncertain situations. We performed experiments based on different challenging scenarios during the platoon’s formation and evolution. At the second level, we incorporate a collision avoidance mechanism using V2V and V2I infrastructures. We used SUMO, Omnet++ with veins for simulations. The results show significant improvement in performance in maintaining traffic flow.

show abstract

Section: Related Workmentioning

confidence: 99%

Traffic Flow Management of Autonomous Vehicles Using Platooning and Collision Avoidance Strategies

et al. 2021

View full text Add to dashboard Cite

show abstract

“…4, the center of the bounding box at t = t 1 + ∆t p can be computed as r n e (∆t p , t 1 ) according to (2). The velocity of the obstacle at t = t 1 + ∆t p , v n e (∆t p ), can be predicted by using (7). Then, the aiming point candidates corresponding to the bounding box at t = t 1 + ∆t p (dotted circle in Fig.…”

Section: B Bounding Tubementioning

confidence: 99%

“…It allows for rapid response to sudden changes in the environment and requires little computational efforts and prior knowledge of the configuration space. From these features, the reactive approach has many benefits in the dynamic environment compared to the proactive planning approach where the UAV incorporates map information to avoid collision with known obstacles in the path planning level, e.g., sampling-based path planning [5], [6] and optimization-based methods [7]- [9]. Vector field approach has been studied as one of the ways of reactive collision avoidance where it utilizes potential functions that consist of repulsive or attractive force fields repelling a UAV from an obstacle or attracting it towards a predefined goal point [10]- [16].…”

Section: Introductionmentioning

confidence: 99%

Reactive Collision Avoidance Algorithm for UAV Using Bounding Tube Against Multiple Moving Obstacles

2020

View full text Add to dashboard Cite

“…Although the conventional method of path following via waypoint tracking works well in most cases without obstacles [24]- [27], there exists a major drawback to this approach. When the surrounding environment contains movable or dynamic obstacles, the generated waypoints become non-reachable if a waypoint is covered by (or close to) an obstacle [28]. As a consequence, when the robot approaches a non-reachable waypoint it enters into a deadlock situation where the simultaneous objectives of reaching the waypoint and avoiding the obstacle contradict each other [29].…”

Section: ) Waypoint Trackingmentioning

confidence: 99%

Collision-Free Trajectory Planning With Deadlock Prevention: An Adaptive Virtual Target Approach

et al. 2020

View full text Add to dashboard Cite

Most human-centred robotic applications require robots to follow a certain pre-defined path. This makes the robot's autonomous movements acceptable and predictable for humans. Planning a trajectory for the robot thus involves guiding it along this desired path. The classical approach of segmenting a path into multiple waypoints and tracking them only works well in environments which are obstacle-free or contain fixed stationary obstacles. Movable or dynamic obstacles that can potentially lie directly on waypoints result in deadlock situations causing the robot to oscillate around the desired waypoint without moving forward. This chapter presents a novel approach for trajectory planning in which an Adaptive Virtual Target (AVT) is formulated that follows the desired path irrespective of surrounding obstacles. The AVT essentially plays the role of a moving reference for the trajectory planner to track. Additionally, the AVT velocity can be adapted such that the robot can catch up in case of deviations from the path due to obstacle avoidance manoeuvres. A model predictive control (MPC) based trajectory planner tracks the AVT and accounts for obstacle avoidance. The proposed approach allows the robot to keep moving towards the goal by preventing deadlocks while simultaneously minimizing deviation from the desired path. Simulations based on a medical X-ray robot are provided to validate the approach.INDEX TERMS Trajectory planning, model predictive control, autonomous robots, deadlock prevention.

show abstract

Two-layer obstacle collision avoidance with machine learning for more energy-efficient unmanned aircraft trajectories

Cited by 43 publications

References 29 publications

Traffic Flow Management of Autonomous Vehicles Using Platooning and Collision Avoidance Strategies

Traffic Flow Management of Autonomous Vehicles Using Platooning and Collision Avoidance Strategies

Reactive Collision Avoidance Algorithm for UAV Using Bounding Tube Against Multiple Moving Obstacles

Collision-Free Trajectory Planning With Deadlock Prevention: An Adaptive Virtual Target Approach

Contact Info

Product

Resources

About