Reciprocal n-Body Collision Avoidance

By Jur van den Berg, Stephen J. Guy, Ming C Lin, Dinesh Manocha. International Symposium of Robotics Research, 2011

1,986 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Mechanical Engineering

View Paper: https://www.semanticscholar.org/paper/69458cdda06755f4fb564ac8a8b9733d3fb21947

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Mechanical Engineering

Citations in the corpus

Decentralized non-communicating multiagent collision avoidance with deep reinforcement learning (2016, methodology): “…let action be the agent’s velocity, a = v; let the observable states be the agent’s position, velocity, and radius (size), so = [px, py, vx, vy, r] ∈ R5; and let the hidden states be the agent’s intended goal position, preferred speed, and heading angle, sh = [pgx, pgy, vpref , θ] ∈ R4.”
Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Prediction (2020, methodology): “ual forces for goal navigation and collision avoidance. Social Force models work well on interaction modeling while performing poorly on trajectory prediction [25]. Geometry based methods, e.g., ORCA [42] and PORCA [34], consider the geometry of the agent and convert the interaction modeling into an”
Optimal Reciprocal Collision Avoidance for Multiple Non-Holonomic Robots (2010, methodology): “Then ORCAi| j = {vHi |(vHi − (v i + cu)) · n ≥ 0} follows as described in [2].”
Unfreezing the robot: Navigation in dense, interacting crowds (2010, background): “We additionally remark that the illustration in Figure 2(b), the crowd experiments catalogued in the research of [8], [9], [7], the multi-robot coordination theorems of [26], [25], and the tracking experiments of [18], [19], [16], all corroborate the argument that autonomous dynamic agents utilize j”
Mobile Robot Path Planning in Dynamic Environments Through Globally Guided Reinforcement Learning (2020, methodology): “Since in Global Re-planning, ECBS and HCA∗, the robot action is not generated online at each step, we only compare the computing time of Discrete-ORCA, PRIMAL and our approach.”
Crowd-Robot Interaction: Crowd-Aware Robot Navigation With Attention-Based Deep Reinforcement Learning (2018, methodology): “In multiagent settings, where the same policy is applied to all the agents, reactive methods such as RVO [5] and ORCA [6] seek joint obstacle avoidance velocities under reciprocal assumptions.”
Deep Reinforcement Learning for Robotics: A Survey of Real-World Successes (2024, methodology): “The policy is preconditioned via finetuning using ORCA (186).”
Core Challenges of Social Robot Navigation: A Survey (2021, methodology): “The literature proposes several techniques to model the crowd’s dynamics, where pedestrian simulation engines are drawn from work in crowd analysis dynamics [103] or computer graphics [59, 123, 261].”
Distributed multi-robot collision avoidance via deep reinforcement learning for navigation in complex scenarios (2020, methodology): “In this way, the sensor-level collision avoidance is more generalized and robust than the agent-level collision-avoid-ance formulation in Chen et al. (2017a,b), van den Berg et al. (2008), and van den Berg et al. (2011a).”
Robot Navigation in Constrained Pedestrian Environments using Reinforcement Learning (2020, methodology): “For example, in the simulation of MESH-worlds, which follows the same principles as the WALLS-environments, the simulation of pedestrians becomes more challenging as the speciﬁc context becomes relevant for sampling the start and end points for the ORCA-driven pedestrians: (e.g. walking from the kit”
PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning (2018, background): “Similarly, but with signiﬁcantly worse performance, ORCA cannot protect against deadlocks and performs very poorly in most scenarios involving more than 16 agents and any obstacles, due to its fully-decoupled, reactive nature.”
Reciprocal collision avoidance with acceleration-velocity obstacles (2011, methodology): “We will combine the approach of [19] with accelerationvelocity obstacles to guarantee collision-free navigation of multiple robots subject to acceleration constraints.”
Algorithms for collision-free navigation of mobile robots in complex cluttered environments: a survey (2014, background): “(RCA) achieve collision avoidance by assuming that each vehicle takes some of the responsibility for each pair-wise conﬂict, with the resulting constraints forming a set of viable velocities from which a selection can be made using linear programming, see, e.g., [258, 283].”
TraPHic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions (2018, methodology): “Examples of methods that explicitly model road-agent interaction include techniques based on social forces [19, 37], velocity obstacles [35], LTA [31], etc.”