MoveIt 2 vs MoveIt Pro 2026: What Qualcomm’s PickNik Deal Means
Qualcomm is buying PickNik, steward of MoveIt. We compare open-source MoveIt 2.15 with MoveIt Pro 10 and explain what changes for ROS 2 manipulation teams.
LeRobot v0.6 vs v0.5: What Changed and How to Migrate
LeRobot v0.6 vs v0.5: world-model policies, GR00T N1.7, lerobot-rollout with DAgger, reward models, six benchmarks, and every breaking change.
World Action Models vs VLAs: Cosmos 3, VLA-JEPA and FastWAM Compared
World action models learn to imagine, then act. How Cosmos 3, VLA-JEPA and FastWAM differ from VLAs, what they cost to train, and where each runs.
Isaac Lab 3.0 vs 2.2: Quaternions Flipped, ProxyArray, and Kit-less Training
Isaac Lab 3.0 flips quaternions to XYZW, replaces torch tensors with ProxyArray, and runs without Isaac Sim. Every breaking change, and how to migrate.
Jetson T3000 vs T4000 vs T5000: Choosing a Thor Module on Bandwidth, MIG, and Power
NVIDIA added T3000 and T2000 to the Jetson Thor line in July 2026. Memory bandwidth, MIG support, networking and power compared for robotics builds.
Nav2 Lyrical vs Kilted: MPPI Trajectory Validation and the New BT Control Nodes
Nav2 for ROS 2 Lyrical adds OptimalTrajectoryValidator for MPPI, per-critic trajectory visualization, and PauseResumeController. What changed and how to migrate.
MIG on Jetson Thor: GPU Partitioning for Mixed-Criticality Robotics in JetPack 7.2
JetPack 7.2 brings MIG to Jetson Thor: two isolated partitions of 12 and 8 SMs. How to split perception, control and GenAI without latency jitter.
Jetson Thor vs Jetson Orin AGX: 2026 Edge AI Upgrade Guide
NVIDIA Jetson Thor vs Jetson Orin AGX in 2026: TOPS, GPU architecture, memory bandwidth, power, price and whether the Thor upgrade is worth it for edge AI.
ROS 2 Kilted Kaiju to Lyrical Luth Migration (2026): What Breaks
Migrating ROS 2 Kilted Kaiju to Lyrical Luth in 2026: breaking changes, ros2_control updates, Zenoh as the Tier-1 RMW and a safe upgrade path.
Diffusion Policy for Robot Manipulation: Imitation Learning (2026)
Diffusion policy explained: how denoising diffusion models generate robot action sequences from demonstrations - visuomotor architecture, training, receding-horizon control and limits.