Robotics Software · Reinforcement Learning · AI/ML
B.Tech Mathematics & Computing, Central University of Karnataka (2023–2027)
I work at the intersection of robotics, reinforcement learning, and intelligent control, with a focus on robot learning, bipedal locomotion, motion planning, and simulation-based control.
My work spans from robot modeling and inverse kinematics to reinforcement learning, trajectory generation, and control, with experience in both simulation and physical robotic systems.
I primarily work with MuJoCo, PyBullet, Gymnasium, and Python, and I am interested in building learning-based robotic systems that can adapt to complex environments and real-world constraints.
Currently exploring: Robot Learning · Reinforcement Learning · Whole-Body Control · MPC · Optimal Control · Sim-to-Real
Robotics & AI
├── Robot Learning
│ ├── Reinforcement Learning
│ ├── Imitation Learning
│ ├── Reward Shaping
│ └── Sim-to-Real
│
├── Humanoid & Biped Robotics
│ ├── Bipedal Locomotion
│ ├── Multi-Skill Locomotion
│ ├── Motion Generation
│ └── Whole-Body Control
│
├── Control & Optimization
│ ├── PD / PID Control
│ ├── Model-Based Control
│ ├── MPC
│ └── Trajectory Optimization
│
└── Planning & Perception
├── Inverse Kinematics
├── Motion Planning
├── A* / Sampling-Based Planning
└── Computer Vision
| Category | Technologies |
|---|---|
| Languages | Python, C++, C, SQL, JavaScript, R |
| Robotics & RL | Gymnasium, Stable-Baselines3, SAC, PPO, TD3, DDPG |
| Control | PD/PID, Model-Based Control, MPC, Whole-Body Control |
| Robotics | Inverse Kinematics, Motion Planning, Robot Dynamics |
| Simulation | MuJoCo, PyBullet, MjLab, Isaac Lab |
| Deep Learning | PyTorch, TensorFlow, Keras |
| Computer Vision | OpenCV, MediaPipe, dlib, ONNX |
| Tools | Git, Linux, TensorBoard, Weights & Biases |
IEEE ICC 2025 | Published
Learning Multi-Skill Locomotion in Underactuated Biped: A Waypoint-Based Reward Shaping Approach
A 6-DOF underactuated biped in PyBullet and benchmarked SAC, TD3 and DDPG on five locomotion tasks (standing, push recovery, walking, uneven terrain, stair descent) using progressive waypoint reward shaping.
- Model-Based Reinforcement Learning
- Multi-Agent Reinforcement Learning
- Whole-Body Control for humanoid robots
- Model Predictive Control
- Sim-to-Real Robot Learning
- Learning-based motion planning
I’m interested in robotics research, open-source robotics projects, and internship opportunities in: Robotics Software · Reinforcement Learning · Robot Learning · ML Engineering
📧 abhaydwivedi10122005@gmail.com
