save

Author: Rwin2

content/authors/admin/_index.md

@@ -14,7 +14,7 @@ organizations:
   - name: Stanford University
     url: https://www.stanford.edu/
 
-bio: "Autonomous systems (RL, perception, decision-making) and LLM-powered scientific agents for multi-omics workflows."
+bio: "Graduate student in Aeronautics & Astronautics at Stanford, working on robot learning and autonomous systems, with a focus on sim-to-real transfer and learned control policies."
 
 interests:
   - Autonomous Robotics

content/project/Language-steered-drones/index.md

@@ -12,7 +12,7 @@ tags:
   - Autonomous Systems
 ---
 
-## Simulated Trajectory
+## Drone navigating to a leaf blower
 <iframe
   width="100%"
   height="420"
@@ -23,7 +23,7 @@ tags:
 
 Developed a vision-language navigation (VLN) policy for autonomous drone flight in photorealistic 3D Gaussian Splatting environments. Given a natural language instruction like "go to the green leafblower," the drone autonomously identifies and navigates to the target — collision-free.
 
-The video shows the drone's onboard view: RGB (left) and semantic similarity field (right) for the query "green and pink leafblower." The system first encodes the language instruction via CLIP embeddings, localizes the target using CLIPSeg semantic segmentation, and generates real-time control commands to navigate through a cluttered indoor environment while avoiding obstacles.
+The video shows the drone's onboard view: RGB (left) and semantic similarity field (right) for the query "green and pink leafblower." In the right view, red indicates high similarity with the query and blue indicates low similarity — the drone navigates towards the high-similarity region while avoiding obstacles. The system first encodes the language instruction via CLIP embeddings, localizes the target using CLIPSeg semantic segmentation, and generates real-time control commands to navigate through a cluttered indoor environment.
 
 The control policy is a lightweight neural network (SqueezeNet Commander MLP) trained via Behavioral Cloning from an ACADOS-based MPC expert. A key contribution is the design and implementation of a full DAgger (Dataset Aggregation) pipeline — including mixed-policy rollouts, expert annotation filtering, iterative retraining with best-model checkpointing, and automated benchmarking — to systematically correct for compounding errors under distribution shift. A second key contribution is the introduction of explicit geometric features — bearing and elevation — extracted from the CLIPSeg heatmap centroid, providing the policy with a direct spatial signal for goal-directed control. This replaces the previous approach where target localization had to be implicitly learned from visual embeddings alone.