From 1ab409446000866ea76c59f15536dbd57ffcd507 Mon Sep 17 00:00:00 2001 From: adaickalavan Date: Fri, 16 Jun 2023 13:19:40 -0400 Subject: [PATCH 1/2] Add gifs for baseline agent in benchmark scenarios. --- docs/benchmarks/driving_smarts_2023_1.rst | 31 +++++++++++++++++++++-- docs/benchmarks/driving_smarts_2023_3.rst | 28 +++++++++++++++----- 2 files changed, 51 insertions(+), 8 deletions(-) diff --git a/docs/benchmarks/driving_smarts_2023_1.rst b/docs/benchmarks/driving_smarts_2023_1.rst index 806d4c2f44..caebba9222 100644 --- a/docs/benchmarks/driving_smarts_2023_1.rst +++ b/docs/benchmarks/driving_smarts_2023_1.rst @@ -45,27 +45,54 @@ below. Any method such as reinforcement learning, offline reinforcement learning, behavior cloning, generative models, predictive models, etc, may be used to develop the policy. -Several scenarios are provided for training. Their names and tasks are as follows. -The desired task execution is illustrated in a gif by a trained baseline agent. +Training scenarios +------------------ + +Several scenarios are provided for training as follows. +The corresponding gifs show the task execution by a trained baseline agent. **Driving SMARTS 2023.1 scenarios** + :scenarios:`cruise_2lane_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_1/cruise_2lane_agents_1.gif + + :scenarios:`cutin_2lane_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_1/cutin_2lane_agents_1.gif + + :scenarios:`merge_exit_sumo_t_agents_1 ` + :scenarios:`overtake_2lane_agents_1 ` + :scenarios:`00a445fb-7293-4be6-adbc-e30c949b6cf7_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_1/00a445fb-7293-4be6-adbc-e30c949b6cf7_agents_1.gif + :width: 75% + :scenarios:`0a53dd99-2946-4b4d-ab66-c4d6fef97be2_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_1/0a53dd99-2946-4b4d-ab66-c4d6fef97be2_agents_1.gif + :width: 75% + :scenarios:`0a576bf1-66ae-495a-9c87-236f3fc2aa01_agents_1 ` **Driving SMARTS 2023.2 scenarios** + :scenarios:`1_to_3lane_left_turn_sumo_c_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_2/1_to_3lane_left_turn_sumo_c_agents_1.gif + :width: 75% + :scenarios:`1_to_3lane_left_turn_middle_lane_c_agents_1 ` + :scenarios:`00b15e74-04a8-4bd4-9a78-eb24f0c0a980_agents_1 ` + :scenarios:`0a60b442-56b0-46c3-be45-cf166a182b67_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_2/0a60b442-56b0-46c3-be45-cf166a182b67_agents_1.gif + :width: 75% + :scenarios:`0a764a82-b44e-481e-97e7-05e1f1f925f6_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_2/0a764a82-b44e-481e-97e7-05e1f1f925f6_agents_1.gif + :width: 75% + :scenarios:`0bf054e3-7698-4b86-9c98-626df2dee9f4_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_2/0bf054e3-7698-4b86-9c98-626df2dee9f4_agents_1.gif + :width: 75% Observation space ----------------- diff --git a/docs/benchmarks/driving_smarts_2023_3.rst b/docs/benchmarks/driving_smarts_2023_3.rst index 365460f3bb..e582a8c06f 100644 --- a/docs/benchmarks/driving_smarts_2023_3.rst +++ b/docs/benchmarks/driving_smarts_2023_3.rst @@ -34,28 +34,44 @@ of steps per episode. Any method such as reinforcement learning, offline reinforcement learning, behavior cloning, generative models, predictive models, etc, may be used to develop the policy. -Several scenarios are provided for training. Their names and tasks are as follows. -The desired task execution is illustrated in a gif by a trained baseline agent. +Training scenarios +------------------ -+ straight_2lane_agents_1 - A single ego must follow a specified leader, with no background traffic. - - .. image:: /_static/driving_smarts_2023/platoon_straight_2lane_agents_1.gif +Several scenarios are provided for training as follows. +The corresponding gifs show the task execution by a trained baseline agent. + :scenarios:`straight_2lane_sumo_agents_1 ` + :scenarios:`straight_2lane_sumo_t_agents_1 ` + :scenarios:`straight_3lanes_sumo_agents_1 ` + :scenarios:`straight_3lanes_sumo_t_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/straight_3lanes_sumo_t_agents_1.gif + + :scenarios:`straight_3lanes_sumo_t_agents_2 ` + :scenarios:`merge_exit_sumo_agents_1 ` + :scenarios:`merge_exit_sumo_t_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/merge_exit_sumo_t_agents_1.gif + + :scenarios:`merge_exit_sumo_t_agents_2 ` + :scenarios:`ff239c9d-e4ff-4acc-bad5-bd55648c212e_0_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/ff239c9d-e4ff-4acc-bad5-bd55648c212e_0_agents_1.gif + + :scenarios:`ff239c9d-e4ff-4acc-bad5-bd55648c212e_agents_1 ` + :scenarios:`ff6dc43b-dd27-4fe4-94b6-5c1b3940daed_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/ff6dc43b-dd27-4fe4-94b6-5c1b3940daed_agents_1.gif + :width: 75% + :scenarios:`ff9619b5-b0c0-4942-b5d8-df6a5814f8a2_agents_1 ` + + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/ff9619b5-b0c0-4942-b5d8-df6a5814f8a2_agents_1.gif + :width: 75% + :scenarios:`ffd10ec2-715b-48af-a89d-b11f79927f63_agents_1 ` + .. image:: https://raw.githubusercontent.com/smarts-project/smarts-project.github.io/master/assets/driving_smarts_2023_3/ffd10ec2-715b-48af-a89d-b11f79927f63_agents_1.gif + :width: 75% + Observation space ----------------- From b990919daa33d6385a1bf535da4a6ee6ef94752d Mon Sep 17 00:00:00 2001 From: adaickalavan Date: Fri, 16 Jun 2023 13:42:20 -0400 Subject: [PATCH 2/2] Rectify spelling. --- docs/benchmarks/driving_smarts_2023_1.rst | 2 +- docs/benchmarks/driving_smarts_2023_3.rst | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/benchmarks/driving_smarts_2023_1.rst b/docs/benchmarks/driving_smarts_2023_1.rst index caebba9222..7034f74a48 100644 --- a/docs/benchmarks/driving_smarts_2023_1.rst +++ b/docs/benchmarks/driving_smarts_2023_1.rst @@ -49,7 +49,7 @@ Training scenarios ------------------ Several scenarios are provided for training as follows. -The corresponding gifs show the task execution by a trained baseline agent. +The corresponding GIF image shows the task execution by a trained baseline agent. **Driving SMARTS 2023.1 scenarios** diff --git a/docs/benchmarks/driving_smarts_2023_3.rst b/docs/benchmarks/driving_smarts_2023_3.rst index e582a8c06f..0a028b7e81 100644 --- a/docs/benchmarks/driving_smarts_2023_3.rst +++ b/docs/benchmarks/driving_smarts_2023_3.rst @@ -38,7 +38,7 @@ Training scenarios ------------------ Several scenarios are provided for training as follows. -The corresponding gifs show the task execution by a trained baseline agent. +The corresponding GIF image shows the task execution by a trained baseline agent. + :scenarios:`straight_2lane_sumo_agents_1 ` + :scenarios:`straight_2lane_sumo_t_agents_1 `