No project description provided
Project description
pedestrian_prediction
Realtime, confidence-varying trajectory prediction for "Probabilistically Safe Robot Planning with Confidence-Based Human Predictions". RSS '18
Installation
git clone https://github.com/sirspinach/pedestrian_prediction.git
cd pedestrian_prediction
pip install -e .
Module contents
pp.mdp
: Data structures and value iteration algorithms for two types of GridWorlds -- a standard GridWorld with only lateral and diagonal movement, and a GridWorldExpanded in which actions operate in "gridless" continuous space but are snapped to the nearest grid cell.pp.inference
: Algorithms for inferring destinations, occupancies, and states given a GridWorld, a list of destinations, and the pedestrian's trajectory so far.pp.util
: Utilities for parsing and trajectory manipulation.pp.plots
: Plot simulated trajectories, overlayed with heatmaps of predicted states. Produce some test plots by executingpython -m pp.plot
. (Requires plotly)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Close
Hashes for pedestrian_prediction-0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6e1ce4435500506060ab20a6434db37a2c25dd996352deb9fcb0c4c0aa0db876 |
|
MD5 | e45f309664df522a2c6da685c1a1bad3 |
|
BLAKE2b-256 | 737dba338fc25682637757a2835b8b4b5c8024d6b3343624d8734788740f117b |