SmartGridDrive: an integrated adaptive Q-learning framework for precision self-parking and reverse navigation in dynamic grid environments—a proof-of-concept study
Revati Raman Dewangan, Deepali Thombre, Vivek Parganiha, Monika Verma, Bhupesh Kumar Dewangan, Amit Pimpalkar, Nilesh Shelke · 2026
Read SmartGridDrive: an integrated adaptive Q-learning framework for precision self-parking and reverse navigation in dynamic grid environments—a proof-of-concept study by Revati Raman Dewangan, Deepali Thombre, Vivek Parganiha, Monika Verma, Bhupesh Kumar Dewangan, Amit Pimpalkar, Nilesh Shelke (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13470224.txt.
Description
Revati Raman Dewangan and colleagues built an adaptive Q-learning parker in 2026 for tight, dynamic grid lots with moving obstacles. Self-parking and reversing strain classic planners, which lack flexibility for shifting slots and hazards. Their hybrid pairs rapidly-exploring random trees for global paths, model predictive control for smooth tracking, and Q-learning for local corrections.
Adaptive exploration decays only when rewards improve, while a Markov look-ahead predicts obstacles and priority updates focus on large-error state-action pairs. In a 10 by 10 simulated grid, success reached 92.4%, collisions just 1.8%, and parking error 6.7 centimetres, beating DDPG, PPO, and DQN baselines. The proof-of-concept backs lightweight tabular learning with prediction for real parking, pending real-world trials.
Catalog record
| CREATOR | Revati Raman Dewangan, Deepali Thombre, Vivek Parganiha, Monika Verma, Bhupesh Kumar Dewangan, Amit Pimpalkar, Nilesh Shelke |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 75 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |