Key Responsibilities:

Reinforcement Learning for Manipulation

• Design and implement deep reinforcement learning (RL) policies for robotic manipulation in dynamic settings

• Develop self-learning and policy optimization techniques to improve decision-making

• Train inverse kinematics (IK) models for real-time, adaptive motion control

Deep Learning for Motor Control & Dexterity

• Build neural network-based control models for grip compliance, force adaptation, and fluid motion

• Use transformer models for intelligent motion sequencing

• Develop Sim2Real pipelines to transfer trained models to physical robots

Motion Planning & Collision Avoidance

• Implement and refine trajectory planning using RRT, PRM, Hybrid-A*, TEB*

• Integrate motion control policies with ROS2 MoveIt! and Orocos

• Enable grasping strategies that adapt to force and handle unstructured environments

Sensor Fusion & Environment Mapping

• Build systems combining data from LiDAR, depth cameras, IMU, and force sensors

• Use Neural SLAM techniques for accurate mapping and object manipulation

• Explore Vision-Language Models (VLMs) to support semantic understanding in robotic tasks

Testing, Simulation & Deployment

• Benchmark model performance against real-world scenarios

• Troubleshoot and refine control pipelines for reliability

• Develop frameworks for Sim2Real validation and deployment

Documentation & Research

• Maintain clear and detailed documentation of models, training processes, and system design

• Stay updated on research in AI, robotics manipulation, and autonomous control systems

Must-Have Skills:

• Strong foundation in Reinforcement Learning, Deep Learning, and trajectory optimization

• Experience with ROS2 MoveIt!, Orocos, NVIDIA Isaac Sim, Groot, and Omniverse

• Hands-on work in Sim2Real transfer and AI-based robotic control

• Familiarity with motion planning algorithms, sensor fusion, and SLAM frameworks