Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
User can set up destination for any agent to navigate on Google Map and learn the best route for the agent based on its current condition and the traffic. Our result is 10% less energy consumption than the route provided by Google map
| Date | Stars |
|---|---|
| 2026-07-31 | 260 |
| 2026-08-01 | 260 |
| 2026-08-02 | 260 |
| 2026-08-06 | 260 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Electrical Vehicle Route Planning on Google Map with Reinforcement Learning More datail please find the **```Paper.pdf```** experiment section. This work is aim to take battery, motor, traffic time, and other factor into account for generating better route (minimize the energy consumption) for the vehicle under certain condition.<br /> Please read carefully through this page before you start running the program. There are many parameters or condition need to be modified in order to fit your environment. <br /> ## Library Requirements with Python 3.6.2: (1) Tensorflow (CPU) (2) numpy (3) pandas (4) haversine (5) time (6) random (7) math (8) requests (9) urllib ## Set Up Google Map API Google Map API is now requiring every search URL contains user's key. Please refer to this link: https://cloud.google.com/maps-platform/ . Enter your billing information, it will not charge you at this moment because you got $200 free search credit. If you exceed the certain amount of search, then you will be charged. You only need the Maps option. <br /> <p align="center"><img src="/image/instruction.JPG" height="60%" width="60%"></p> You will get a key for example **```AZuesYuds12_dsakd23456sdeHf```**.Then the searching URL of (latitude, Longitude) = (40.468254,-86.980963) will become: https://maps.googleapis.com/maps/api/geocode/json?address=40.468254%2C-86.980963&key=AZuesYuds12_dsakd23456sdeHf <br /> The result of the URL (with your correct key) will be like the following figure.<br /> <p align="center"><img src="/image/googlemapresult.JPG" height="40%" width="40%"></p> There are three places in the **```Environment.py```** where you need to add the key like the figure below. I've put these fake key in the code for you. What you need to do is replace these fake key with your own key.<br /> <p align="center"><img src="/image/googleURL.JPG" height="60%" width="60%"></p> You can learn how to interact with google map api (geocode/elevation/directions) correctly in **```Environment.py```**. <br /> Notice that my research and result was conducted in 2017 in Purdue ME linux server. During that time, I'm free to do the search without paying any fee (Sometimes my search would stop because Google map server would block my ip because I've done too much query. Wait for 5 to 10 min or sometimes even longer such as 1 to 2 hours, and the server would unlock my ip.) I'm sure Google Map server is not going to block your ip any more because they will charge your credit card directly. ## How can you get started Download all the python file as following: **```main.py```**, **```Environment.py```**, **```DoubleDQN.py```**, **```battery.py```**, **```motor.py```** <br /> * enter your start position in **```main.py```**: name of position or geocode (lat, lng)<br /> * enter your destination position in **```main.py```**: name of position or geocode (lat, lng)<br /> * enter the length of each step (step_length) in **```main.py```** line 78, higher but less accurate (ex: 1000m takes less time to train compare to 100m)<br /> * enter how many episode you want to train in **```main.py```**<br /> * There are many file and folder path name need to be changed in **```main.py```** to fit your condition.<br /> * Make sure that you can access to the internet and google map during the training process. Notice that your server will be blocked if your access to google map API exceed the limit in 24 hour and your program will be stucked. In the **```main.py```** we have implemented a mechanism to make your program sleep for a while when you over query the data. If you can get the full access to the Google map API, then you can remove the ```sleep``` command and make the learning process faster. (This rule might be changed with new rule adopted by Google Map Server. If so, please disable all the sleep command ~ edit on April, 2019)<br /> * If you never train any model yet, then you need to make **```load_model == False```** in the line 141 of **```main.py```** .
Excerpt of 18,225 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:ebc69d201efe286f, topic:reinforcement-learning, name:reinforcement learning