completed codes of the lecture note of probabilistic robotics ("詳解確率ロボティクス" in Japanese)

• basic methods
  • mcl.ipynb: Monte Carlo localization
  • kf.ipynb: Kalman Filter
• advanced methods
  • kld_mcl.ipynb: MCL with KLD (Kullback–Leibler divergence) sampling
  • adaptive_mcl.ipynb: Adaptive MCL
  • sensor_reset_mcl.ipynb: MCL with sensor resetting
  • expansion_reset_mcl.ipynb: MCL with expansion resetting
  • expansion_sensor_reset_mcl.ipynb: MCL with sensor resetting and expansion resetting
  • occlusion_free_mcl.ipynb: MCL with a non-Gaussian likelihood function

• fastslam1.0.ipynb: FastSLAM 1.0
• fastslam2.0.ipynb: FastSLAM 2.0
• graphbasedslam.ipynb: graph-based SLAM
• misc
  • graphbasedslam_logger.ipynb: logger for graph-based SLAM

• dynamic_programming.ipynb: value iteration
• policy_agent.ipynb: agents with the result of value iteration

• q.ipynb: Q-learning
• sarsa.ipynb: simple sarsa
• nstep_sarsa.ipynb: n-step sarsa
• sarsa_lambda.ipynb: sarsa($\lambda$)

• qmdp.ipynb: Q-MDP value method
• pfc.ipynb: weighted Q-MDP value method (probabilistic flow control)
• amdp.ipynb: value iteration with an augumented MDP (an example of corstal navigation by N. Roy in another environment and setting)
• amdp_policy_agent.ipynb: agents with the result of amdp

• gauss_gamma.ipynb: Bayes inference with Gauss-gamma distribution
• variational_inference.ipynb: variational inference