Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments
This paper presents a novel motion planning framework for mobile robots operating in dynamic and uncertain environments, with an emphasis on accurate trajectory prediction and safe, efficient obstacle avoidance. The proposed approach integrates search-based planning with deep learning techniques to...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/13/7551 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849319992874500096 |
|---|---|
| author | Tengfei Liu Zihe Wang Jiazheng Hu Shuling Zeng Xiaoxu Liu Tan Zhang |
| author_facet | Tengfei Liu Zihe Wang Jiazheng Hu Shuling Zeng Xiaoxu Liu Tan Zhang |
| author_sort | Tengfei Liu |
| collection | DOAJ |
| description | This paper presents a novel motion planning framework for mobile robots operating in dynamic and uncertain environments, with an emphasis on accurate trajectory prediction and safe, efficient obstacle avoidance. The proposed approach integrates search-based planning with deep learning techniques to improve both robustness and interpretability. A multi-sensor perception module is designed to classify obstacles as either static or dynamic, thereby enhancing environmental awareness and planning reliability. To address the challenge of motion prediction, we introduce the K-GRU Kalman method, which first applies K-means clustering to distinguish between high-speed and low-speed dynamic obstacles, then models their trajectories using a combination of Kalman filtering and gated recurrent units (GRUs). Compared to state-of-the-art RNN and LSTM-based predictors, the proposed method achieves superior accuracy and generalization. Extensive experiments in both simulated and real-world scenarios of varying complexity demonstrate the effectiveness of the framework. The results show an average planning success rate exceeding 60%, along with notable improvements in path safety and smoothness, validating the contribution of each module within the system. |
| format | Article |
| id | doaj-art-0d97c497d0f24e09b4904cbd4a52c3e8 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-0d97c497d0f24e09b4904cbd4a52c3e82025-08-20T03:50:16ZengMDPI AGApplied Sciences2076-34172025-07-011513755110.3390/app15137551Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic EnvironmentsTengfei Liu0Zihe Wang1Jiazheng Hu2Shuling Zeng3Xiaoxu Liu4Tan Zhang5Sino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaSino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaSino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaSino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaSino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaSino-German College of Intelligent Manufacturing, Shenzhen Technology University, Shenzhen 518118, ChinaThis paper presents a novel motion planning framework for mobile robots operating in dynamic and uncertain environments, with an emphasis on accurate trajectory prediction and safe, efficient obstacle avoidance. The proposed approach integrates search-based planning with deep learning techniques to improve both robustness and interpretability. A multi-sensor perception module is designed to classify obstacles as either static or dynamic, thereby enhancing environmental awareness and planning reliability. To address the challenge of motion prediction, we introduce the K-GRU Kalman method, which first applies K-means clustering to distinguish between high-speed and low-speed dynamic obstacles, then models their trajectories using a combination of Kalman filtering and gated recurrent units (GRUs). Compared to state-of-the-art RNN and LSTM-based predictors, the proposed method achieves superior accuracy and generalization. Extensive experiments in both simulated and real-world scenarios of varying complexity demonstrate the effectiveness of the framework. The results show an average planning success rate exceeding 60%, along with notable improvements in path safety and smoothness, validating the contribution of each module within the system.https://www.mdpi.com/2076-3417/15/13/7551motion planningdynamic obstaclesGRUsLSTM |
| spellingShingle | Tengfei Liu Zihe Wang Jiazheng Hu Shuling Zeng Xiaoxu Liu Tan Zhang Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments Applied Sciences motion planning dynamic obstacles GRUs LSTM |
| title | Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments |
| title_full | Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments |
| title_fullStr | Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments |
| title_full_unstemmed | Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments |
| title_short | Adaptive Motion Planning Leveraging Speed-Differentiated Prediction for Mobile Robots in Dynamic Environments |
| title_sort | adaptive motion planning leveraging speed differentiated prediction for mobile robots in dynamic environments |
| topic | motion planning dynamic obstacles GRUs LSTM |
| url | https://www.mdpi.com/2076-3417/15/13/7551 |
| work_keys_str_mv | AT tengfeiliu adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments AT zihewang adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments AT jiazhenghu adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments AT shulingzeng adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments AT xiaoxuliu adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments AT tanzhang adaptivemotionplanningleveragingspeeddifferentiatedpredictionformobilerobotsindynamicenvironments |