Skip to content

Commit

Permalink
doc: English page update (#47)
Browse files Browse the repository at this point in the history 
* doc: Fixed the EN top page

* doc: add image files

* delete images folder

* doc: add link page

* doc: delete "Page Links" in intro

* doc: change customize

* doc: add online

* doc: change rule

* doc: change setup

---------

Co-authored-by: toki-1441 <hogemoge@example.com>
  • Loading branch information
@toki-1441
toki-1441 and toki-1441 authored Dec 20, 2023
1 parent 0002a4d commit 810c6ad
Show file tree
Hide file tree
Showing 7 changed files with 178 additions and 221 deletions.
242 changes: 60 additions & 182 deletions documentation/docfx_project_en/customize/index.md
View file

Large diffs are not rendered by default.

98 changes: 74 additions & 24 deletions documentation/docfx_project_en/index.md
View file
Original file line number Diff line number Diff line change
@@ -1,8 +1,4 @@
# Japan Automotive AI Challenge 2023 (Simulation)

![AIchallenge2023-Simulation](../images/top/aichallenge-racing-top.jpeg)

We are hosting the Japan Automotive AI Challenge 2023 (Simulation), a motorsport-themed software development competition. Participants will develop autonomous driving software based on [Autoware.Universe](https://github.com/autowarefoundation/autoware.universe), and integrate it into a racing car that runs in the End to End simulation space (AWSIM). The goal is simple, win the race while driving safely!
## Japan Automotive AI Challenge 2023 (Simulation)

## Competition Concept

Expand All @@ -12,48 +8,102 @@ As a new initiative to discover and develop engineers who will lead the automoti

We aim to build a framework that contributes to the further development of the automobile industry, starting from this competition, which is conducted through industry-government-academia cooperation.

## Overview
![AIchallenge2023-Simulation](../images/top/aichallenge-racing-top.jpeg)

## Japan Automotive AI Challenge 2023 (Simulation) Goals

Motorsports, which test the driving performance of automobiles in extreme conditions, have contributed to the improvement and development of automobile performance by creating technical challenges. The 2023 Autonomous Driving AI Challenge (Simulation) is aimed at contributing to the development of autonomous driving technology that can safely and quickly operate in an end-to-end (E2E) simulation environment.

Autonomous driving in racing involves a variety of technical elements, such as control technology that takes into account lateral slip, air resistance, and vehicle response characteristics, as well as tracking and behavior prediction of other vehicles. This tournament provides participants with an opportunity to develop autonomous driving software with strategic driving planning and advanced vehicle control functions that incorporate these elements, and to share their knowledge. We expect that engineers who have acquired new technologies through this Challenge will contribute to technological innovation in the automotive industry.

## Japan Automotive AI Challenge 2023 (Simulation) Summary

<video width="1040" height="585" controls>
<source src="../images/top/racing_sample.mp4" type="video/mp4">
Your browser does not support video tags.
</video>

In this tournament, it is possible to run in the video mentioned above using the software that is distributed in advance. Using this software, please make improvements in line with the challenges described below.

### Automated Driving Technology Competition and Special Awards

Participants will aim to win the race while safely driving with autonomous driving software based on [Autoware.Universe](https://github.com/autowarefoundation/autoware.universe). In addition, to promote the growth of the community of this tournament and to share knowledge that is useful for participants, the following special awards will be established. Please see [the overview page](https://www.jsae.or.jp/jaaic/2023_simulation.php) for more details.

1. The Community Contribution Award, which honors those who have contributed to the OSS as volunteers or to the expansion of the community.

2. The Sponsor Award, which is voted on by platinum sponsors to encourage active initiatives through AIChallenge.

![Start](../images/top/aichallenge-racing-start.jpeg)

The race in the End to End simulation environment is themed, aiming to contribute to the development of safe and robust autonomous driving technology for high-speed driving.

Autonomous driving in races involves various technical elements such as skidding, air resistance, vehicle response characteristics, tracking other vehicles, and behavior prediction. This competition offers participants an opportunity to work on developing autonomous driving software that incorporates these elements, featuring strategic driving planning and advanced vehicle control functions. We hope that engineers who learn new technologies through this "challenge" will contribute to technological innovation in the automotive industry.
### Driving Course and Rivals

| <img src="../images/top/sim_monza.png" width="640"/> | <img src="../images/top/vector_map_monza.png" width="640"/> |
| :-------------------------------------------------: | :--------------------------------------------------------: |
| MonzaTrack at AWSIM | MonzaTrack at vector map |

The race course for this tournament will be the [MonzaTrack](https://www.google.com/maps/search/MonzaTrack). It is about 5.8 kilometers long and has a complex shape, including long straights and hairpin turns. Participants will race on this circuit-themed course along with multiple rival vehicles. As rival vehicles will accelerate, brake, and change course in response to your autonomous vehicle, strategic path planning is required.

### Vehicle to be used

In this competition, we are using vehicles that will be used in the Indy Autonomous Challenge.

| <img src="../images/top/sim_car.png" width="640"/> | <img src="../images/top/indy_car.jpg" width="640"/> |
| :-----------------------------------------------: | :------------------------------------------------: |
| simulation car | indy car |


## The 3 Challenges
### Challenge 1: Perception
### Challenge 1: Robust Perception

In the first challenge, participants are required to predict the future behavior of other vehicles based on the vehicle recognition data provided by the simulator. Autonomous vehicles require a high level of safety, and it is essential to detect and predict the movements of other vehicles that exist in the surrounding 360 degrees through sensor data in order to avoid collisions. In this competition, we provide accurate data (ground truth) as vehicle recognition information so that participants can focus on behavior prediction.

![Challenge 1: Robust Perception](../images/top/aichallenge-racing-perception.jpeg)
| <img src="../images/top/aichallenge-racing-perception.jpeg" width="640"/> | <img src="../images/top/sim_routing.png" width="640"/> |
| :----------------------------------------------------------------------: | :---------------------------------------------------: |
| sensing and perception | perception at simulation |

Specifically, the following points can be improved

- Other vehicle tracking: Participants use sensor data to track the location and movement of other vehicles. This process includes analyzing information from the sensors and accurately identifying vehicles in the vicinity.
- Behavior prediction: Based on the data from tracked vehicles, the future behavior is predicted. This prediction is essential for collision avoidance and the development of safe driving strategies.

Safety posits always highest priority in autonomous driving. The first step to avoid collisions is to recognize other vehicles from sensor data and predict their movements. The first challenge is the development of fast and robust recognition functions.

### Challenge 2: Strategic Route Planning

![Challenge 2: Strategic Route Planning](../images/top/aichallenge-racing-planning.jpeg)

The second challenge is strategic route planning. To drive faster than others, it is necessary to process the ever-changing surrounding environment and constantly pursue the optimal route.
The second challenge is strategic path planning. In order to drive faster than other vehicles, it is necessary to process the constantly changing surrounding environment using the recognition results and always pursue the optimal route.

| <img src="../images/top/aichallenge-racing-planning.jpeg" width="640"/> | <img src="../images/top/sim_routing.png" width="640"/> |
| :--------------------------------------------------------------------: | :---------------------------------------------------: |
| real environment planning | route planning |

### Challenge 3: Control in High-Speed Driving

![AIchallenge2023-Racing](../images/top/aichallenge-racing-control.jpeg)
Specifically, the following points can be improved

- Dynamic situation response: The situation during a race is constantly changing. It is required to respond to these factors quickly and effectively, such as how other cars move and how the road surface changes.
- Pursuit of the optimal route: In order to drive faster than other vehicles, it is necessary to find and pursue the optimal route based on this information. This includes things like how to take curves, the timing of acceleration and deceleration, and judging the chance of overtaking.

### Challenge 3: Control in High-Speed Driving

To win a race, control technology that realizes the optimal route planning is essential. The third challenge is the development of control modules that enable driving along strategic and complex routes while driving in high-speed.

## Competition Description
The third challenge is the development of a control module that can handle high speeds and allow driving along strategic and complex routes. In order to win a race, it is essential to have control technology that can realize the optimal path planning. In this competition, accurate data (true values) are provided as self-localization results so that participants can focus on vehicle control.

### Racing Course and Rivals
| <img src="../images/top/aichallenge-racing-control.jpeg" width="640"/> | <img src="../images/top/sim_control.png" width="640"/> |
| :-------------------------------------------------------------------: | :---------------------------------------------------: |
| high speed control | high speed control at simulation |

![Start/Finish Straight](../images/top/aichallenge-racing-course.jpeg)
Specifically, the following points can be improved

The course of this competition is approximately 5.8km long and has a complex shape including long straights and hairpin curves. You will drive on this circuit-themed course with several rival cars. The rival cars will accelerate and decelerate or change courses according to your autonomous driving car, so strategic route planning is required.
- High-speed control: During a race, especially when driving at high speeds of around 200 km/h, vehicle control, including gear shifting, is very important. Driving at high speeds is very different from driving at low speeds, so more precise control is required.
- Vehicle interface development: In this challenge, a specialized control module is required to be developed to enable driving at high speeds. This module must properly control the vehicle's accelerator, brake, and steering using a simulator that accurately simulates actual vehicle behavior.

### Evaluation Criteria

The goal of this competition is to win the race while driving safely. Therefore, the performance of your autonomous driving software will be evaluated from the following three perspectives. Details on scoring and rules are due to be published in this documentation later.
This tournament aims to quickly and safely reach the goal. Therefore, the driving of the participants' autonomous driving software will be evaluated based on the following three aspects. For details on scoring and rules, please visit the rules page.

1. Driving Distance - Reaching the goal of the course
2. Safety - Driving without collisions or deviations from the route
3. Speed - The time to the finish line
1. Distance - Reach the goal of the course.
2. Safety - Drive without collision or route deviation.
3. Speed - Drive to the goal in a short time.

### Development Environment

Expand Down
2 changes: 0 additions & 2 deletions documentation/docfx_project_en/intro/toc.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,4 @@
href: index.html#about-awsim
- name: Related Documentations
href: index.html#related-documentations
- name: Page Links
href: index.html#page-links
homepage: intro/index.html
4 changes: 2 additions & 2 deletions documentation/docfx_project_en/online/index.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,8 +48,8 @@
- autoware.log
2. Upload to the Online Environment
<img src="../images/online/siteImage.png" width="100%">
&emsp; Access the [online environment](https://aichallenge.synesthesias.jp) and upload the aichallenge_submit.tar.gz file you created in step 1 by clicking "Choose File." Once the upload is complete, the source code will be built, and the simulation will be run in that order.
<img src="../../images/online/siteImage.png" width="100%">
&emsp; Access the [online environment](https://aichallenge.synesthesias.jp) and upload the ```aichallenge_submit.tar.gz``` file you created in step 1 by clicking "Choose File." Once the upload is complete, the source code will be built, and the simulation will be run in that order.

* If the process finishes successfully, "Scoring complete" will be displayed, and you can download the result.json file. The distance points and time will be displayed on the ranking.
* When a scenario finishes running but no scores are output due to launch failure, etc., the result will be "No Result" and will not be used as the final time. If you see "No Result" even though the local run is working properly, there may be an internal error on the server side, so please upload the file again. Please contact us if the message is displayed repeatedly.
Expand Down
14 changes: 8 additions & 6 deletions documentation/docfx_project_en/rule/index.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,14 @@ Certainly! Here's the translation of the provided text in the same format:
| 50 | N/A | 00:10 | N/A | 5 |
| 10 | N/A | 00:00 | N/A | 6 |

### Goal Position
- The completion of the run (reaching the goal) is when the following goal points are crossed.
```yaml
goal.position.x: 21912.0
goal.position.y: 52267.5
```
- A goal can be set by using goal_pose_setter.
### Penalty
&emsp;Violations such as deviating from the course or colliding with other vehicles are penalized. The penalties for violations are:
Expand All @@ -39,12 +47,6 @@ Below are the violations and their corresponding penalties:
- Deviating more than 5m from the course boundary for more than 10 seconds
- Being more than 100 seconds away from the course boundary
- Completion of the run is considered when crossing the following goal point.
Sample can be found at `docker/aichallenge/aichallenge_ws/src/aichallenge_submit/autoware_micro/goal_pose_setter/config/default_goal_pose.param.yaml` (Note: yaml file is not specified in the launch file)
```yaml
goal.position.x: 21912.0
goal.position.y: 52267.5
```
**Major Violation (Penalty: 10 seconds/occurrence):**
Expand Down
2 changes: 1 addition & 1 deletion documentation/docfx_project_en/setup/index.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
<br>

> [!REGISTER]
> こちらから参加登録!
> Register here!
> [https://www.jsae.or.jp/jaaic/en/index.php](https://www.jsae.or.jp/jaaic/en/index.php)
<br>
Expand Down
37 changes: 33 additions & 4 deletions documentation/docfx_project_en/toc.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -28,14 +28,43 @@
href: online/index.html
- name: Teams
href: ../teams/index.html
#- name: Customizing Autoware
# href: customize/index.html
- name: Customizing Autoware
href: customize/index.html
- name: FAQ
href: FAQ/index.html
- name: Blog Posts
dropdown: true
items:
- name: Advent Calender
href: https://qiita.com/advent-calendar/2023/jidounten-ai
- name: FAQ
href: FAQ/index.html
- name: Past Event Blog Posts
dropdown: true
items:
- name: "(1) Try to run Autoware"
href: https://qiita.com/h_bog/items/86fba5b94b2148c4d9da
- name: "(2) Successfully avoiding the first obstacle"
href: https://qiita.com/h_bog/items/a36df3042a342bd2a0c5
- name: "(3) Success in avoiding the second obstacle"
href: https://qiita.com/h_bog/items/849a4241cdef257f3b3a
- name: "(4) All obstacle avoidance achieved! (All parameters disclosed)"
href: https://qiita.com/h_bog/items/8bce776fef0a39676c46
- name: "(5) Restructuring of development environment (extra)"
href: https://qiita.com/h_bog/items/4ed94ef6256ac5ae9c03
- name: "(6) Using Autoware-Mini"
href: https://qiita.com/h_bog/items/075543f30f9b3432fa1d
- name: "(7) External module introduction (failure)"
href: https://qiita.com/h_bog/items/bcfd5648afa8b6dee5b3
- name: "(8) Introduction of home-grown packages"
href: https://qiita.com/h_bog/items/368543a4b10b356db0e5
- name: "(9) Creation of Obstacle Avoidance Module①"
href: https://qiita.com/h_bog/items/c84c4886c0ba0e6c1724
- name: "(10) Creation of Obstacle Avoidance Module②（Success？）"
href: https://qiita.com/h_bog/items/a2c2cbdcec6a852bfcd5
- name: "(11) Improved obstacle avoidance module①"
href: https://qiita.com/h_bog/items/4160e396fc5def3c8383
- name: "(12) Improved obstacle avoidance module②"
href: https://qiita.com/h_bog/items/c7ffd9d51e01adf8e021
- name: "(13) Improved obstacle avoidance module③"
href: https://qiita.com/h_bog/items/770e4e2437a963082b12
- name: To Japanese Page
href: ../index.html

0 comments on commit 810c6ad

Please sign in to comment.