Moving in a Crowd: Safe and Efficient Navigation among Heterogeneous Agents

Authors: Julio Godoy, Ioannis Karamouzas, Stephen J. Guy, Maria Gini

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Simulation experiments performed in many scenarios demonstrate that an agent using our approach computes safer and more time-efficient paths as compared to those generated without our inference approach and a state-of-the-art local navigation framework. and Third, we experimentally validate our approach and show that it leads to safer and more efficient motions in a variety of scenarios as compared to a state-of-the-art collision avoidance framework [van den Berg et al., 2011], and to planning without estimation of the models and goals.
Researcher Affiliation Academia Julio Godoy, Ioannis Karamouzas, Stephen J. Guy, and Maria Gini Department of Computer Science and Engineering University of Minnesota 200 Union St SE, Minneapolis MN 55455 {godoy, ioannis, sjguy, gini}@cs.umn.edu
Pseudocode Yes A pseudocode of our approach can be seen in Algorithm 1. and Algorithm 2 shows the model selection procedure. and Algorithm 3 shows the pseudocode of the planning process using an action set, Actions, for the entire time horizon T.
Open Source Code No The paper does not provide concrete access to source code for the methodology described, such as a specific repository link or an explicit code release statement.
Open Datasets No The paper does not provide concrete access information (specific link, DOI, repository name, formal citation with authors/year, or reference to established benchmark datasets) for a publicly available or open dataset.
Dataset Splits No The paper describes simulation scenarios but does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or testing.
Hardware Specification Yes All simulations ran in real-time on a Core i7 CPU with 8 GB of memory.
Software Dependencies No The paper states, "We implemented our inference approach and our two planning methods, Vel Plan and Social Plan, in C++," but does not provide specific version numbers for C++ or any other ancillary software dependencies or libraries used.
Experiment Setup Yes We updated the position of every t = 0.1s, which was also the planning time limit. We empirically set the values of the time horizon T to 50 timesteps, γ to 0.1%, and used a σ value of 10 4 (Eq. 1), as they provided the best performance compared to larger values.