4D imaging radar in 2026: overview
4D imaging radar estimates an object's range, azimuth, elevation and velocity, creating a richer scene representation than conventional radar. Its combination of depth and Doppler information can support perception when lighting or weather limits cameras, while radar’s sparse, noisy returns still make object classification and precise localization difficult. The current technical push is to extract more useful structure from radar data and combine it with complementary sensors.
Recent work is tackling those limits through learning, fusion and mapping. In September 2026, Shanghai JiaoTong University researchers introduced Diff4RadBEV, using conditional diffusion to enhance sparse, noisy radar bird’s-eye views. A multimodal traffic-scene dataset developed with Tongji University and Zhejiang University also appeared that month. Other directions include radar-camera fusion, self-supervised odometry and calibration; papers from August 2026 address cross-modal learning from LiDAR and radar-based mapping. These efforts reflect a shift from treating radar as a standalone detector toward a complementary perception and localization sensor.
- 1Learning from sparsity: Diffusion enhancement and learned perception target radar’s sparse, noisy point clouds rather than treating them as complete scene descriptions.
- 2Fusion is accelerating: Multisensor fusion is gaining momentum as researchers combine radar with cameras and LiDAR to improve perception in difficult conditions.
- 3Beyond road vehicles: Off-highway machinery is an emerging application, where radar’s ability to perceive through some obstructions can complement other sensors.
Top organizations working on 4D imaging radar
Ranked by Linknovate activity score · last 24 monthsThe activity score adds up the relevance of every patent, publication, grant and news item linking an organization to 4D imaging radar. Showing 7 of 127 organizations.
| # | Organization | Approach | Signals (24 mo) | Momentum | Why it matters |
|---|---|---|---|---|---|
| 1 | 
NXP Semiconductors
Corporate · Netherlands |
General | PUB 3 · NEWS 41 | -43% | Develops automotive radar semiconductor components, including radar transceivers used in vehicle sensing systems. |
| 2 | 
Infineon Technologies
Corporate · Germany |
General | PUB 3 · NEWS 33 | -67% | Develops automotive radar chips and sensing components that support radar-based driver-assistance systems. |
| 3 | 
Continental AG
Corporate · Germany |
General | NEWS 34 | -58% | Supplies automotive radar sensors and driver-assistance systems, an established route for integrating radar perception into vehicles. |
| 4 | 
Robert Bosch GmbH
Corporate · Germany |
General | NEWS 38 | -48% | Develops automotive radar sensors and driver-assistance systems for vehicle detection and safety functions. |
| 5 | 
Arbe Robotics Ltd.
Startup · Israel |
MIMO array architectures | NEWS 23 | -65% | Develops high-resolution automotive radar for ADAS and autonomous driving, including 4D imaging radar systems. |
| 6 | 
Tongji University
Research · China |
Multisensor fusion | PUB 21 | -25% | Recently, researchers developed a method for constructing multimodal 4D imaging radar datasets for 3D traffic scenes. Other work addresses radar-camera calibration and feature fusion for 3D object detection. |
| 7 | 
Denso Corporation
Corporate · Japan |
General | NEWS 30 | -50% | Develops automotive sensing and driver-assistance systems, including radar-based vehicle perception. |
| 8 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 9 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 10 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
Startups to watch and leading research centers
Startups to watch 1 total
Bitsensing Inc.South KoreaDevelops automotive and smart-city radar systems. Its AIR4D imaging radar is designed for autonomous-vehicle fleets.
Leading research institutions 109 total
- Tongji UniversityChinaRecently, researchers developed a method for constructing multimodal 4D imaging radar datasets for 3D traffic scenes. Other work addresses radar-camera calibration and feature fusion for 3D object detection.
Zhejiang UniversityChinaRecently, researchers contributed to work on a multimodal 4D imaging radar dataset for 3D traffic scenes. Its research also includes low-cost single-chip automotive radar and learned 3D object detection from 4D radar.
Shanghai JiaoTong UniversityChinaRecently, researchers proposed Diff4RadBEV, a conditional-diffusion method to enhance sparse, noisy 4D radar bird’s-eye-view data. Other work covers point-cloud registration, classification, and radar-based SLAM.- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
Where 4D imaging radar innovation happens
Share of documents with at least one organization from each country.
Open the interactive map with all 127 organizationsMain technical approaches
Last 24 months| Approach | What it is | Documents | Leading organizations |
|---|---|---|---|
| MIMO array architectures | MIMO arrays synthesize a larger virtual aperture for angular resolution, trading hardware and signal-processing complexity against finer spatial detail. | 37 |
Boeing Company, BAE Systems, Safran |
| Beamforming and angle estimation | Beamforming and angle estimation infer target direction from received signals, balancing angular precision against computation and sensitivity to interference. | 38 |
NXP Semiconductors, BMW AG, Infineon Technologies |
| Signal processing and point clouds | Signal processing converts radar measurements into usable detections and point clouds, with the trade-off that sparse, noisy returns limit scene detail. | 232 |
Tongji University, NXP Semiconductors, Shanghai JiaoTong University |
| Learned radar perception | Learned perception models extract objects and motion from radar data, but their performance depends on representative training data and robust generalization. | 164 |
Tongji University, Zhejiang University, University of Science and Technology of China |
| Multisensor fusion | Multisensor fusion combines radar with cameras or LiDAR, improving complementary perception while adding calibration and synchronization challenges. | 153 |
Robert Bosch GmbH, Tongji University, Continental AG |
Emerging trends
Last 12 months vs previous 36 monthsTerms whose frequency in the last 12 months grew the most compared with the previous 36 months.
Funding and investment
Recent funding rounds
FACTS reports $6,229,902 across 14 grants since 2020. The listed funding rounds are not clearly attributable to 4D imaging radar, so they are not presented as radar-specific investment.
Patent landscape
Top patent holders · last 24 monthsRecent developments
Latest news, duplicates removed- Aug 2026Huawei-backed Maextro launched the V800 luxury MPV in China on August 5, with a top trim price of 1.016 million yuan (149,700 USD). The brand also introduced the smaller V680, priced at 648,000 yuan (95,400 USD). The…CarNewsChina.com
- Jul 2026Changan’s high-end electric brand Avatr obtained an L3 autonomous driving test license, aiming at the commercialization of advanced intelligent driving. At the same time, Avatr submitted its listing application to the…CarNewsChina.com
- May 2026Xiaomi is turning up the heat in the performance EV space with two new additions to its YU7 luxury SUV range — the YU7 Standard and the range-topping YU7 GT. And it’s the GT that’s grabbing headlines. The YU7 GT has…Indian Autos Blog
- May 2026Xiaomi launched a new entry-level YU7 “Standard Edition” today priced at RMB 233,500 (~$32,400) — undercutting the Tesla Model Y by RMB 30,000 ($4,350) while delivering 643 km of range versus the Model Y’s 593 km. CEO…Electrek
- Month 20XXHidden developmentAvailable to registered users with alerts enabled.
Frequently asked questions about 4D imaging radar
What companies are working on 4D imaging radar?
FACTS identifies 18 companies working in 4D imaging radar, including NXP Semiconductors, Infineon Technologies, Continental AG, Robert Bosch GmbH and Arbe Robotics Ltd. Their activity spans radar components, vehicle sensors and high-resolution imaging radar. Texas Instruments, Denso Corporation and ZF Friedrichshafen AG are also among the organizations active in the field.
Which 4D imaging radar startups are most promising in 2026?
FACTS identifies Bitsensing Inc. as a startup active in 4D imaging radar, while Arbe Robotics Ltd. is another startup developing high-resolution automotive radar. Bitsensing develops automotive and smart-city radar systems and describes its AIR4D product as an imaging radar for autonomous-vehicle fleets. Arbe’s work includes applications beyond passenger vehicles, including off-highway machinery.
Which universities and research centers lead 4D imaging radar research?
Tongji University, Zhejiang University and Shanghai JiaoTong University are prominent research organizations in FACTS. Their work includes multimodal radar datasets, learned object detection and radar bird’s-eye-view enhancement. Other active groups include Tsinghua University, Beijing Institute of Technology and Southeast University, with research spanning fusion, odometry and radar perception.
How much funding has 4D imaging radar received?
FACTS reports $6,229,902 across 14 grants since 2020. This figure is the available grant total; it does not establish that every grant exclusively funded 4D imaging radar. The listed funding-round records are broad organizational financing entries and are not clearly attributable to radar development, so they should not be treated as a radar-specific investment total.
Who holds the most 4D imaging radar patents?
Sandvik Mining And Construction leads the patent-holder list in FACTS with 4 patents in the dataset. Nanjing Movelaser Co. follows with 2. Other listed holders include Shanghai JiaoTong University, Jiangsu University, University of Toledo, Xidian University and Hyundai Motor Company, each with 1 patent.
Is 4D imaging radar patent activity growing?
FACTS records 29 patents in the 24-month window, and its patent-growth indicator shows a decline. Publication activity is moving in the opposite direction: FACTS reports 214 publications last year, with publication growth of 39.9%. This contrast suggests recent activity is more visible in research than in patent filings.
What are the latest developments in 4D imaging radar?
Recent papers highlight radar enhancement and sensor fusion. In September 2026, Shanghai JiaoTong University researchers presented Diff4RadBEV, a conditional-diffusion method for enhancing radar bird’s-eye views; a multimodal traffic-scene dataset involving Tongji University and Zhejiang University also appeared that month. August 2026 work explored cross-modal learning from LiDAR, radar odometry and radar-camera mapping.
What are the applications of 4D imaging radar?
FACTS lists driver assistance and vehicle safety, autonomous driving and robotaxis, off-highway machinery, and robotics and industrial automation as applications. Research also explores maritime perception, autonomous ship navigation, human action recognition and mapping. In mining, 4D mmWave radar is being studied with LiDAR to address perception blind spots caused by equipment occlusion.
Which countries lead 4D imaging radar?
The United States leads the country list in FACTS with 1,488 documents, followed by China with 1,007 and Germany with 723. These figures describe the dataset’s country associations, not exclusive national capabilities. Other countries with notable activity include the United Kingdom, Japan, South Korea, France and Israel.