Loading...
Please wait, while we are loading the content...
Similar Documents
Predicting Gaits Based on Prior Experience Using the Monte Carlo Method
| Content Provider | Semantic Scholar |
|---|---|
| Author | Fahle, Hans Fredrik |
| Copyright Year | 2017 |
| Abstract | Robots are becoming a valuable part of the society. An important reason behind this is that the robots are becoming more capable of operating without human interactions. Today the robots are capable of moving around with certain limitations. This gives the opportunity to increase the value that robots have by improving their ability to move around independently. This study aims to determine if prediction, utilizing prior experience, can be used to find the most suitable gait for traversing a certain type of terrain. The robot is a four-legged robot with five different gait configurations. The accumulation of the experience data is done by making the robot traverse five different terrain types using the five different gaits. For this process the Monte Carlo method was used. The experience data accumulated is then used to predict the most suitable gait when traversing each terrain type on a test terrain. To evaluate the performance of the prediction approach a second approach is implemented. The second approach is a traditional approach which evaluates the stability of the robot while traversing a test terrain. The final result in this thesis compare the prediction approach with the traditional approach. This comparison gives an indication of how well the prediction approach performs. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://www.duo.uio.no/bitstream/handle/10852/57438/Predicting_gaits_based_on_prior_experience.pdf?isAllowed=y&sequence=1 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
